Adobe Interviews

I recently went through the interview process for an Adobe MTS-2 role. After an initial call from HR, she scheduled my first two technical rounds for the following week.

Round 1: DSA

This round focused on Data Structures and Algorithms. I was asked two specific tree-related problems. The first involved finding the minimum time for a message to reach all employees in a hierarchical organization with the CEO as the root, where messages are passed in parallel from parent to child. The second was a variation where messages were passed sequentially, with a time increment for each successive child (e.g., CEO at t=0, first child at t=1, second at t=2). I provided solutions for both questions and coded them out, and the interviewer seemed satisfied with my performance.

Round 2: LLD+DSA

This round combined Low-Level Design with some DSA elements. The main task was to implement an an in-memory file system, including core functions like ls, mkdir, addFile, and getFile. I successfully coded a working solution for this. Following this, the interviewer asked how I would modify the code if the data type of the file content changed (I had initially assumed it to be a string). This part was essentially a question about Java generic datatypes. I cleared both these rounds and received confirmation for the next two rounds.

Round 3: LLD

This round was dedicated to Low-Level Design. I was tasked with designing a warehouse system where a warehouse contains multiple spaces, and these spaces are then rented to firms. The design needed to handle optimal space allocation and be able to process orders for items within the warehouse. Unfortunately, this round did not go as well as the previous ones, and I believe this round ultimately led to my rejection.

Round 4: HM round

The Hiring Manager round was less technical. He mostly asked about some of my resume projects and then delved into a few theoretical Springboot questions, as Springboot was part of the job description. Finally, he asked some behavioral questions. This round lasted about 40 minutes and, overall, I felt it went quite well.

Verdict: I received a rejection a few days later.

I recently had an interview experience with Adobe for a CS1 role. I have 4 years of experience and was interviewing for their ALM Org. The process involved three distinct rounds, each testing different facets of my technical and behavioral skills.

Round 1 (Technical)

Round 2 (Technical)

Round 3 (System Design & Behavioral)

I recently had my interview experience for the Adobe MTS 2 role, which spanned a total of six rounds and took two months to complete.

ROUND 1: Technical Interview

This round involved a medium-level question primarily focused on the Disjoint Set Union (DSU) data structure, followed by several follow-up questions to delve deeper into my understanding.

ROUND 2: Technical Interview - Toughest Round

This was by far the toughest round for me. It covered a wide array of in-depth topics:

• Extensive questions on Multithreading.
• Detailed discussion on B+ trees.
• Problems related to Bipartite graphs.
• Questions on Virtual Destructors in C++: what they are, why they're needed, and how they function.
• The infamous Diamond Problem in C++.

There was a significant number of questions from these topics, pushing my knowledge boundaries.

ROUND 3: Technical Interview

This round also had a strong focus on C++ concepts. Key areas included:

• Numerous C++ questions.
• A problem based on the Two-Pointer technique.
• Questions regarding Connected Components in graphs.
• One Low-Level Design (LLD) question: designing a garbage collector in C++.

ROUND 4: Technical Interview

Similar to previous rounds, this one also heavily featured C++ questions. A lot of focus was placed on Smart Pointers. Additionally, I faced one question that required a Backtracking approach.

ROUND 5: Hiring Manager Round

Having successfully navigated four technical rounds, this round was purely behavioral. The Hiring Manager posed many tricky questions, which I tried to answer as diplomatically as possible. I must admit, I fumbled quite a bit due to panic, but I tried my best to maintain composure.

ROUND 6: Director Round

In stark contrast to the previous rounds, especially Round 2, this was the easiest. I had a very pleasant and insightful discussion with the Director, who seemed like a genuinely nice person – a rare trait to find at such a senior position. After this discussion, I felt quite confident about receiving an offer.

True to my feeling, after two days, the HR informed me that I had successfully cleared all rounds! I was absolutely thrilled to receive the offer.

Round 1: Data Structures & Algorithms

I was presented with two questions. The first problem asked me to minimize the sum of array elements after performing a division operation K times. I solved this LeetCode easy question using a PriorityQueue, achieving a time complexity of O(N log N) and space complexity of O(N).

The second question was a slight variation of the classic Knapsack Problem, which I approached with a 2D DP solution and some optimizations, resulting in O(N) time and O(1) space complexity.

Round 2: Low-Level Design (LLD)

In this round, I was asked to design a configuration management service, detailing the class designs, attributes, methods, and appropriate design patterns.

Round 3: Low-Level Design (LLD)

This round also focused on LLD and included two questions. The first was to design an auto-complete search system that predicts Top K elements based on search frequency. I tackled this problem using a Trie data structure with some variations, identifying it as a LeetCode hard question. The second question involved designing the Java implementations for Linux commands like mkdir, ls, and delete for files and directories. I used a Tree Data Structure for this, but unfortunately, I could only complete one function within the given time.

Despite my efforts, the final verdict was a rejection.

As a Data Scientist/MLE with 3+ years of experience and a B.Tech from IIT (non-circuital branch), I went through a structured interview process at Adobe. The first round was an end-to-end design question focusing on image + text (document-type) data, requiring a complete pipeline to predict aspect ratio from the given data/document. The second round included a Python coding question followed by a deep dive into LLMs, transformers, and attention mechanisms. The third round was a managerial discussion on projects, teamwork, and behavioral questions. The fourth round involved a system design question to create a color suggestion system, along with team and work-related discussions. I was offered the position and will share compensation details in another post soon.

My interview process for a Senior Android Design Engineer role on the Spectrum Design Team at Adobe consisted of three rounds.

Round 1: Hiring Manager Round
This round was with a very friendly Hiring Manager. We discussed my previous experiences, with a particular focus on Jetpack Compose and my experience with Design Systems. The manager was especially interested in the interoperability between View systems and Compose.

Round 2: Android Coding
In this coding round, the interviewer asked me to implement a Compose function. The question revolved around drawing objects on the screen and ensuring they were retained correctly during orientation changes. Key concepts discussed included remember, rememberSaveable, and Custom Saver.

Round 3: Android Coding
For this coding round, I was allowed to use Android Studio with auto-complete, which was helpful. The core problem was about retrieving data from a given endpoint and then efficiently populating that data into a list view.

Throughout the coding rounds, the focus was heavily on Android Jetpack Compose View concepts such as Stable and Immutable states, as well as performance considerations for views, including efficient data handling, pagination, and lazy loading. There were no DSA (Data Structures & Algorithms) rounds.

I had my first interview round for an MTS2 C++ position at Adobe on August 26th. The interview felt positive, and I believe I performed well during the session. However, since then, there has been no communication or notification from their side regarding the outcome or next steps.

I applied for the CS1 position through Adobe's career site. About 15 days later, I received a call from HR to schedule an interview.

Round 1 (DSA)

Round 2 (DSA)

Round 3

I applied for several roles at Adobe, primarily MTS 2, and received a callback from a recruiter after about a month. The interview process at Adobe is structured with sequential rounds.

Round 1: DSA, System Design, OOPs, DBMS

My first round focused on Data Structures & Algorithms, System Design, Object-Oriented Programming (OOPs), and Database Management Systems (DBMS). I was asked a LeetCode medium-hard question:

• Find Median from Data Stream: I clearly built the solution using two priority queues and explained my approach, writing pseudo code.

Then, the interviewer delved into various system design questions covering scaling, performance, concurrency, load balancers, CAP theorem, and DBMS concepts like locking, SQL vs NoSQL, normalization, OS-level locks, and atomicity. I also had to implement an LRU cache, again providing pseudo code. This round lasted approximately 70 minutes.

I received an email for Round 2 the same day.

Round 2: DSA, CI/CD, AWS, Kubernetes, Docker, Kafka

The second interviewer was very approachable. They posed a medium DSA question that was quite lengthy to read but essentially involved backtracking with proper constraint handling. I had to write the full code and test it with sample cases. My second question was LeetCode 825 (Friends Of Appropriate Ages), which I solved and demonstrated with provided inputs. Following this, we discussed my experiences with Jenkins, CI/CD pipelines, build processes, production tools, AWS, Kubernetes, Docker, scaling, and Kafka.

Two days later, I received an email for Round 3. By this point, I had already secured a better offer, so I approached this round very calmly, mostly out of curiosity.

Round 3: Hiring Manager (HM) - Projects, Behavioral, Release Engineering

This was a hiring manager round where I was thoroughly grilled on my projects and past experiences. The questions covered ownership, challenges encountered, end-to-end project involvement, leadership skills, innovation, voluntary acts, and debugging skills.

The HM then informed me that the role was primarily for a Release Engineer with a strong focus on Jenkins, which significantly diminished my interest in joining Adobe. Despite this, I maintained a facade of interest, and they proceeded with Jenkins-specific questions.

• Job Scheduling (Two Schedulers): Given an array of job times, I was asked to determine the minimum time to complete all jobs using two schedulers, where each scheduler can perform one task at a time. I solved it and explained my approach.
• Job Scheduling with Dependencies: The next question involved tackling job scheduling when jobs have dependencies. I discussed using topological sort and how the presence of a loop would make the problem unsolvable, outlining my strategy to handle this.

The interviewer seemed pleased with my responses.

After another two days, I received an email stating I had cleared all rounds, with only a director round remaining (described as similar to a standard HM round). However, I ultimately declined the offer as I decided not to proceed further.

Overall, my experience was positive. The HR team was very accommodating and adjusted interview times promptly upon my request. Interviewers were friendly and provided ample time for me to think through and articulate my approaches.

Hi everyone, I recently completed the interview process for the Computer Scientist II role at Adobe and wanted to share my experience. I have around 8 years of experience in backend development, working with technologies like Java, Spring Boot, Kafka, Redis, and AWS. I’ve been leading a backend team for the past couple of years and was referred internally for the Noida office.

Interview Process Overview:

Round 1: Online Coding Assessment This was hosted on HackerRank. There were two medium-level DSA questions (one based on sliding window and one involving graph traversal) and a short design-based coding task to implement an LRU cache. I focused on writing clean and modular code and completed all three with optimal or near-optimal approaches.

Round 2: Live Coding + CS Fundamentals This round started with a coding problem that was a variation of the merge intervals question. I walked through my approach, discussed edge cases, and iterated with the interviewer. We also had a quick discussion on system fundamentals — threads vs processes, memory paging, database indexing, and REST API basics. The conversation was friendly and didn’t go too deep; it felt more like a warm-up for the design-focused rounds ahead.

Round 3: Low-Level Design (LLD) Prompt: Design a Notification System supporting Email, SMS, and Push. I approached this by defining clear interfaces for Notifier and NotificationChannel, and used the strategy pattern to support multiple channels. I created a central NotificationService that handled orchestration. We discussed extending it with support for retries, rate-limiting, and asynchronous execution using Java's ExecutorService. I also explained how I’d unit test the logic and handle dependency injection. The interviewer appreciated the clarity and asked how I’d integrate a message queue like Kafka down the line.

Round 4: High-Level Design (HLD) Prompt: Design a scalable file storage system (similar to Dropbox or Google Drive Lite). I started by defining basic functional requirements such as file upload/download, link sharing, and optional versioning. I broke the system into services: API Gateway, Auth Service, File Upload Service, Metadata Store, Link Generator, and Object Storage. We discussed chunked uploads, deduplication using content hashes, and storing metadata in RDS with Redis caching. We also went over monitoring, retry handling, consistency guarantees, and security via signed URLs. The interviewer pushed me on trade-offs and scaling decisions but kept it collaborative throughout.

Round 5: Hiring Manager Round This was more conversational and focused on my past projects, how I handle production issues, mentoring responsibilities, and experiences with cross-functional teams. I spoke about a few systems I had built end-to-end, decisions I had to defend in design reviews, and my role in driving architectural changes. The hiring manager also explained how Adobe’s teams are structured, what the CS2 role typically involves, and what they expect from someone at this level — primarily strong ownership, technical depth, and mentoring junior engineers.

Offer and Outcome: I received the offer about a week later. The role was for Computer Scientist II at the Noida office.

If you’re preparing for a CS2-level role at Adobe or a similar company, I’d recommend spending most of your time on LLD and HLD. Practice open-ended design questions, review your own past system decisions, and be ready to talk through them in detail. DSA is still useful but not the main focus.

Hope this helps someone who’s preparing.

For those curious — I got referred through a combination of platforms like LinkedIn and https://boostmyreferral.com. If you're actively looking, exploring referral networks definitely helps open doors, especially for roles with internal referral bias.

Some interview experiences help you grow. Others help you develop patience. This one helped me question reality.

It began like any other opportunity — a call from HR for the Computer Scientist 1 role. She outlined a standard process: 4–5 technical rounds, including interactions with a Manager and a Director.

I was optimistic. I shouldn’t have been.

Round 1: DSA + C++ Fundamentals

The interview was rescheduled by an hour.

It began with some classic C++ theory:

Difference between pointer and reference

What is a stack overflow? Can it happen without recursion?

What is heap memory?

**DSA Problems:

1. House Robber – Leetcode**

2) A string-to-integer conversion problem with edge-case handling:

Ignore leading/trailing spaces

Only valid if all characters are digits or a leading '-'

Handle overflows (return INT_MAX if invalid)

Cannot assume fixed integer size (compilor-dependent, some use 2 Bytes some use 4 Bytes of integer)

I solved both, clean and clear.

And then came the plot twist: “Can you please click on your Chrome search bar and show me your recent search history?”

Yes, really.

Apparently, cracking DSA problems live isn't enough. The real test is whether you've Googled too much lately. I half-expected a pop quiz on my Amazon order history next.

Round 2: A Masterclass in Mismanagement

This round would’ve been funny — if it wasn’t real. It was rescheduled four times, each for a different reason:

1. Interviewer had a health issue , rescheduled.

2. I joined on time, waited 30 minutes, no one showed ,HR was unreachable via call or email Hours later: “It’s rescheduled again.”

3. Interviewer was “busy” , rescheduled again

4. HR wasn’t sure whether the interviewer was busy or not , rescheduled again 🤯

It felt like I was part of a mockumentary on corporate disarray.

After all that chaos, the actual round started.

Quick experience chat, then two DSA problems:

1) Count of Smaller Numbers After Self — solved via stack

2) https://leetcode.com/problems/container-with-most-water/description/ — I used the optimal two-pointer approach

What happened next made me question if I was being tested on code — or on my ability to deal with intellectual stubbornness.

The interviewer seemed dead set on a stack-based solution . I explained my approach using multiple test cases, walked through logic, and offered alternate examples. Still, the same questions came again — not with curiosity, but resistance.

It was clear: I hadn’t misunderstood the problem — I had outpaced her expectations. And maybe, unintentionally, her ego.

She closed with a lecture on how "interviewers know all approaches" and that "candidates should just listen."

I waited to ask follow-up questions, but she cut the call with a cold “bye” and muted herself. I was left speaking into silence — and then, the screen.

Post-Interview: The Final Vanish

Days later, a rejection email quietly landed. I reached out for feedback — the HR literally cut my calls everytime(called multiple HRs). I emailed again. No response. Just… silence.

No clarity. No closure. Just a strange emptiness — the kind only a truly broken process can leave behind.

Background I’m currently a frontend developer at a fintech startup with 2 years and 10 months of experience. I recently interviewed at Adobe for the CS-1 role. I cleared all five rounds with positive feedback, but since CS-1 requires a minimum of 4 years of experience, I was offered the MTS-2 role instead, which I accepted.

Round 1: JavaScript Fundamentals This round focused on JS core concepts like closures, the event loop, async behavior, and context binding. Coding questions:

1. Implement a debounce function
2. Flatten a deeply nested object using dot notation
3. Compose async functions with chaining support (composeAsync) We also talked about memory leaks in large React apps, garbage collection, and retry strategies for flaky APIs. The feedback was positive.

Round 2: React Implementation I was asked to build a nested comment section with support for replies, like/dislike, auto-saving drafts, and keyboard accessibility. Follow-up questions included:

• Custom hook for infinite scroll with cursor-based pagination
• Optimizing re-renders using React.memo, useCallback, and virtualization
• Handling large lists and conditional rendering for better performance The interviewer appreciated the way I structured and explained the solution.

Round 3: Frontend System Design I had to design a content management dashboard with real-time updates and access control. We discussed:

• Component-level architecture
• Polling vs WebSocket trade-offs
• Caching strategies
• List virtualization
• Handling offline mode and sync conflicts
• Tracking frontend performance using tools like Sentry and Web Vitals Feedback from this round was strong.

Round 4: Manager Round The first part of the round focused on my current work, team dynamics, and expectations from Adobe. DSA questions:

1. Search in a 2D sorted matrix
2. Follow-up: Replace the found element and restore sorted order with minimal operations
3. Hard DP problem: Given a grid with obstacles, find the number of unique paths from top-left to bottom-right, with at most one obstacle allowed to be removed I solved the matrix problem completely and walked through the DP approach verbally for the unique paths question. The manager was satisfied with my reasoning.

Round 5: Director Round We had a casual discussion about my projects, interests, and why I wanted to join Adobe. DSA problem: Reverse nodes in k-group in a linked list Follow-up: Modify to solve using iterative method with O(1) space Bonus DP question (asked verbally): Given a string, find the minimum number of cuts needed to make all substrings palindromes (Palindrome Partitioning II) I gave a detailed breakdown of the approach using bottom-up DP and space optimizations. The director appreciated the clarity.

Final Outcome All rounds were positive. Since I had just under 3 years of experience, Adobe moved ahead with the MTS-2 offer instead of CS-1. I accepted it as the team, scope of work, and growth path felt right.

Round 1(DSA):

This round was primarily focused on DSA. Started with intro and questions on projects from resume. Interviewer asked 1 medium Leetcode problem with working code on Hackerrank. Coded with optimal approach & edge cases. Then asked minimum cost to cut sticks problem, for which I explained and told that this is standard DP problem. So he changed question to create binary tree using preorder and inorder traversal. Explained approach but didn't ask for code. Then was asked 1 more leetcode medium question. Explained approach but wasn't able to write complete working code due to time constraints but interviewer was satisfied. In end, he asked on how to handle multi-concurrency in java.

Round 2(Design):

LLD round on how to design Train ticket booking(IRCTC). Several use cases like book trains between stations, find trains, how to assign seats, routes, letting users book tickets if user dropped in between and seat is vacant post certain station etc. Majorly started from DB schema and how overall usecses would work via sql queries.

Round 3(Design):

LLD round on how to create Checkout system after adding items to cart. Should handle authentication and authorization for all API calls. Started with entities and was asked for Producton ready code. Wrote modular services with abstraction, interfaces. Used patterns, asked for more patterns so that this could be improved. Explained on that part. Interviewer was satisfied. Didn't ask to run the code.

Round 4(Managerial Round): The interview started with the introduction. We had discussions around my project and several questions on it. Then it was HLD and LLD round on how to schedule meetings on google calender. Input: N users having vacant meeting slots. Schedule a meeting so that all/max users are available. It was more of how to approach the problem. Started with usecases, Schema design, SQL queries to power usecases and mix of Merge intervals algorithm.

Round 5(Director round): Started with intro. Discussed how previous interviews went, discussed questions on projects. Behavioural questions like why Adobe, challenging thinks I worked upon etc. Went pretty well.

Overall it was a good interview experience. The overall process took 3 weeks. Accepted the offer.

Overall Verdict: Selected

Feel free to add comments if you want to ask something. Will add compensation details soon.

Background - Currently working as a frontend engineer at one of India’s leading quick-commerce companies with 2 years and 11 months of experience.

I interviewed at Adobe for a CS-1 (Computer Scientist 1) role. I cleared all five rounds with positive feedback, but due to my years of experience falling short of their CS-1 requirement (minimum 4 years), I was eventually offered the MTS-2 (Member of Technical Staff - 2) role instead.

Round-1
(JS Fundamentals and Coding)
This round focused on core JS concepts:
Topics: Closures, this context, call/apply/bind, event loop, etc.
Problem 1: Based on closure
Problem 2: Related to prototype chaining
We also discussed web security and frontend performance optimizations.
Feedback: Positive

-> It was a positive feedback

Round-2
(React machine coding round)
I was asked to:
Build a custom hook to support pagination with hasNext, cursor, pageSize, etc.
Implement infinite scrolling where nextPage is triggered on scroll end.
I completed it within 45 minutes. The interviewer was happy with the result.
We then had a casual chat about Adobe’s culture, which was really helpful and positive.

-> It was again positive feedback

Round-3
(Web system design round)
This was divided in two halfs
In first half i need to design a app which will sort of config driven
Had discussion on BFF layer, pros and cons, how we can achieve, how we can optimize that by using caching and all

Second half, she asked me to design facebook feed page
I approached it step-by-step:
1. Functional & non-functional requirements
2. MVC architecture
3. BFF and API interfaces
4. Real-time data with sockets or polling
5. Rendering optimizations
6. Accessibility concerns

The interviewer was from the Acrobat Android team, so we went deep into the BFF integration aspect.

-> Again a positive feedback


Round-4
(Manager Round)
first half -> More about my work, which team i am going to join, hows the culture and what is my expectations
second half -> 1 Dsa problem
Search in 2D sorted matrix
was able to solve it in 15 mins, Then he extended the problem by saying: after finding the element, I need to replace it with a new number, and after that, the sorted order of the matrix must be maintained with the minimum number of operations.
I was able to discuss about how i can approach it optimially, he was satisified by the verbal approach and asked me to dry run once

-> Positive feedback and hr asked me if i am available for the director round on the same day, i was ready so i said yes


Round-5
(Director round)
It was pretty chill round
Discussion on my projects, why i want to join adobe, what is motivation at work daily, and he tell me more about the team and the people
After that he asked me one DSA problem
Reverse linked list in k group, i wrote down the code in js


The HR called the next day with highly positive feedback.
However, they informed me that I’d be offered the MTS-2 position instead of CS-1 due to my YOE being just under 3 years (CS-1 requires 4+ years minimum).

Though initially it felt odd (as I had interviewed for CS-1 and cleared all rounds), I was still happy to accept the offer and excited to be part of Adobe. 🎉

I appeared for an interview with Adobe few days ago.

The interviewer herself has no idea how the question has to be solved.
She kept on asking random questiond around the logic and even though i pointed out multiple times she keeps coming with invalid input for the question.

I gave her brute force which was working for all the scenarios except her invalid ones, with O(n*n) TC.
Then i optimized the solution to O(n), for which also she was stuck with invalid test cases only.

I clearly told her multiple times this was not within the given questions constraint and in order to handle those scenarios the whole question approach will be changed.

Still, suggested her the variation that will be possible to accomodate her changes and told the approach with the optimized tc.

Didn't expected such an experience from Adobe.

Question was some variation of https://leetcode.com/problems/jump-game-ii/description/

You have been given an array and you are at index 0.

print the maximum jumps too reach the end

Ex: - [2, 3, 1, 1, 4] -> 2->3->4
Ex: - [5, 1, 1, 4, 2, 2, 1]-> 5->4->1

TC: - O(N)
sc: - O(N) for vector

Exp : 3 Years

Application Process: Applied on Instahyre

Virtual Interview -

Round 1- DSA

Question-1

https://leetcode.com/problems/majority-element/description/

Question-2

https://leetcode.com/problems/count-unreachable-pairs-of-nodes-in-an-undirected-graph/

I had 2 more rounds scheduled later for that day but didn't happen as this round was elimination round. Most optimal space & time solution was expected. Code should be runnable and pass the test cases.

Verdict : Rejected

College : Tier 1.5

Current Experience : 4 years and 10 months at MNC

Team I was interviewing for - Illustrator

Round 1 Technical Feedback - positive:

This round focused mainly on DSA and C++. For the DSA part pseudo code was needed not the working code.

DSA questions were as follows -

• Buy and Sell Stock - and its further variations Buy and Sell Stock II, Buy and Sell Stock III.
• House Robber - and its variation House Robber II, .
• Single Number - and its variation Single Number III (difficult to arrive at if you haven't practiced bit manipulation. I had done this question earlier, so knew the concept behind it.)

Then there were C++ questions related to mutex locks and a code snippet to print a specific order of output from 2 different functions in a multi-threaded environment. I had to apply the mutex locks to do this. This round went very well.

Round 2 Technical Feedback - positive:

Some questions regarding unique_ptr and move semantics. 1 questions regarding the output of a code snippet. I had to fix it. It was related to unique_ptr and move semantics Some questions on memory managerment.

• Find Cousins of a node in a binary tree - I was able to propose 1 pass solution. Mix of pseudo and working code was required.

Then the inteviewer asked to design an elevator system (more on the logic side rather than high level design) which is time efficient as well as energy efficient. I gathered the requirements. Wrote downt the use cases/scenarios. Explained them.

This round also went pretty well.

Round 3 Manager(Techno Managerial) Feedback - positive:

Asked about my projects a bit. Problems faced, how did you solve them, etc. There were some C++ questions - regarding memory management, etc. Asked me one DSA question-

• Check if a tree is a BST or not.

Then there were C++ questions regarding singleton class. Write a class where each object has a unique ID. Copying/assigning an abject should also lead to unique ID. Could not implement it fully.

This round went well apart from the last question.

Round 4 Director(Techno Managerial):

The round was supposed to take place at 10 AM. The interviewer didn't join till 10:15. I called the recruiter, I was told that the interviewer won't be able to join. As I was leaving for office, I recieved a call that can you join now, interviewer is available. I joined at 10:35, the interviewer joined at 10:45 or 10:50.

After that everything was in a rush. Asked about my projects a bit. Problems faced, how did you solve them, etc.

Then asked the 3 priests and 3 devils puzzle. He just wanted the steps. Asked me to solve it and then "just tell me the steps, don't tell me the result after every step".

Then he showed me a feature of the Illustrator software. Asked me to implement it. I was able to write something. I wasn't sure how much was it correct. The interviewer seemed a bit satisfied with my approach though.

Verdict

Didn't receive any feedback for a week. The recruiter kept on saying that she hasn't received any feedback. After one week I called again, the recruiter said that yout feedback is positive but the team has identified a candidate internally (WTAF). I was so pissed. The recruiter then said that she'll try to align me with some other team as my feedback had been positive. Only a 30 minutes director round would be needed more if she finds a team. It has been a week since then. She said that she doesn't have any updates as of yet and she'll update me when there is any change.

TL;DR:

Adobe made me go through 5-6 hours of interview process just to tell that although my feedback was positive but they have internally identified a candidate. The recruiter is trying to find some other team but hasn't received any update(I don't know how much truth is there to this but I'm now just waiting).

Too much frustrated as this has happened to me twice now in a span of two weeks. The other experience - https://leetcode.com/discuss/post/6811285/oracle-epm-senior-software-developer-ban-lsd3/

My interview journey at Adobe for the MTS-2 position in Noida was a comprehensive process, culminating in an offer after about three weeks.

Round 1: Data Structures & Algorithms

This round was primarily focused on DSA. I was first asked to solve the Maximum Product Subarray problem. After successfully solving it, the interviewer asked for a slight modification: to also return the actual subarray itself, not just the maximum product. Following this, I was tasked with implementing an LRU Cache in O(1) time in Java. The interviewer then extended this by asking me to design the code such that a user could select different caching algorithms, like LRU or LFU, with maximum code reuse.

Round 2: Java & Low-Level Design

This round specifically tested my Java proficiency and LLD skills. I started with a discussion on OOPS concepts, followed by questions on abstract classes, their implementation, and the key differences between abstract classes and interfaces. I was also asked to implement an interface. I then tackled some complex code snippets related to abstract classes and interfaces, which truly tested my core Java understanding. A specific question was whether two interfaces can be implemented by an abstract class. We then moved on to implementing polymorphism and discussing various ways to create threads in Java (extending Thread, implementing Runnable, and Callable).

Towards the end of this round, I was given an LLD snapshot with API contracts and had to design the models, their data types, and pseudo-code for the repository, service, and controller layers. The interviewer then asked me to adapt my design to use SQL for data implementation, moving away from in-memory storage. This led to questions about different types of SQL Joins and their distinctions.

Since my previous role involved Node.js, I was also asked to explain how a server runs in Node.js and the principles of the HTTP protocol. The interviewer seemed satisfied at the end of this extensive round.

Round 3: Managerial Round

This round began with introductions and quickly moved into a deep dive into my past projects. We discussed my specific roles and implementations. The interviewer was particularly interested in understanding my contribution in terms of business impact, asking if I measured my work in GMV (Gross Merchandise Value) terms. I was also asked about my involvement in open-source projects. Standard behavioral questions were part of this round. Finally, there was a discussion about potential use cases for AI in my current role.

Round 4: Directorial Round

This was arguably the most challenging round. The director rigorously questioned me about a project involving MySQL to Cassandra migration. We delved into the differences between these two databases, their respective benefits and shortcomings, and the various types of keys used in Cassandra. The interviewer went deep into each project I mentioned, even asking about the underlying mathematics behind some of the functions I had implemented.

The entire interview process took approximately two weeks. In the third week, I received the call informing me of my selection and offer.

Round 1: This round was primarily focused on DSA.

1. Trapping Rain Water: https://leetcode.com/problems/trapping-rain-water/
2. Given an OBJ_ID(key) which is really long number and cannot be represented in primtive types like long long, int, etc. Implement a data structure which performs get, insert and delete in constant time. Keep in mind that don't use unordered_map since in worst case it can go till O(n).

Note: In second question, I was stuck on the unordered_map approach and after a lot of discussion with the interviewer why it would not work and the possible scnearios. The trie approach hit me and I discussed that with the interviewer and he was satisfied.

There were also very basic questions around C++ fundamentals.

Round 2: At the start of the interview, I was bombarded with a lot of questions around data types, in which part of the memory are static, global variables stored, smart and raw pointers comparison. Some coding questions:

1. Given a number n, find the number of trailing zeroes at the end od n factorial.
2. Give a working code where using a raw pointer is leading to a memory leak and then using a smart pointer fixes it.
3. https://leetcode.com/problems/coin-change/description/

Round 3(Managerial Round): The interview started with the introduction. We had discussions around my project and what was my role in it. Then we jumped onto the coding questions:

1. https://leetcode.com/problems/intersection-of-two-linked-lists/ (Immediately gave the approach and interviewer asked not to code it and jumped to next question)
2. The scenario is something similar to above and the point is now you have large number of linked lists and they can merge at any point with any linked list. Keep in mind that interviewer was completely against using any extra space, by which I mean is O(n) and time Complexity of O(n) was expected. Return the vector<vector> which merge at some point. (PS: I think this was a tricky question if not seen beforehand, solution would be to store the tail and the correspoding list in an unordered_map. I initially started thinking into a graphical/tree approach. Took me a while to come back and think simple). After this, there was discussion around AI which was something general.

Round 4(Directorial Round): In depth project discussion based on the technicalties and how specific things were handled. Design the Acrobat Desktop file system. I asked some clasrifying questions and scope of the problem. Was able to quickly implement it and it ended fairly early.

Overall it was a good interview experience. The overall process took a week for me.

Overall Verdict: Selected

Hi community,
From long time I have been reading leetcode interview experience section but i guess this is the time to contribute.

College: Tier 1
Current Organsation: PBC(Startup)
YOE: 2 years 1 month

Recently an Adobe recruiter reached out to me and for MTS-2 role and she mentioned that all the rounds will be in single day only as it is a hiring drive.

Round 1: Problem Solving(DSA)

Given an array of words and an array of sentences, calculate how many sentences can be created by replacing any word with one of its anagrams.
Note:
•	Two words are said to be anagrams of each other if one can be created by rearranging the letters of another word, using all the original letters exactly once.
wordSet = ['listen', 'silent', 'it', 'is']
sentences = 'listen it is silent', 'silent it is silent'
wordSet = ['cat', 'act', 'is']
sentences = 'cat is tac', 'act tac is cat'
Write the countSentences function in the editor below.
countSentences has the following parameters:
•	string wordSet[n]: an array of strings
•	string sentences[m]: an array of strings
Returns:
•	int[]: an array of m integers that denote the number of sentences that can be formed from each sentence

Question 2:

Imagine you are designing the organizational hierarchy of a large company. In this structure, each node represents a department, and the edges represent direct managerial relationships between departments.
	• The company consists of n departments, labeled from 0 to n - 1.
	• You are given n - 1 manager-subordinate relationships in the form of an array edges, where each element [a_i, b_i] indicates that department a_i is the direct manager of department b_i, or vice versa. This structure forms a tree (i.e., there are no circular managerial chains).
Task:
Your goal is to choose the most efficient department to serve as the central management hub for the entire company. The height of the hierarchy represents the longest managerial chain (in terms of reporting lines) from the central management department to the most distant department.
In a well-designed organizational structure, you want the central management department to be as close as possible to all other departments, minimizing the distance (in terms of reporting lines) to any department.
Objective:
You need to identify which departments can serve as the best central management hubs. These are the departments that, when selected as the central hub, result in the shortest longest reporting chain to the most distant department. This ensures that communication and decision-making processes are as efficient as possible.
Output:
Return a list of all the departments (node labels) that can act as the best central management hubs for the company. The answer can be returned in any order.
Input: n = 4, edges = [[1,0],[1,2],[1,3]]
Output: [1]
Input: n = 6, edges = [[3,0],[3,1],[3,2],[3,4],[5,4]]
Output: [3,4]

I answered both of them and within 10 minutes i got email to appear for

Round 2: Problem Solving(DSA)

Question 1:

Imagine you are designing the organizational hierarchy of a large company. In this structure, each node represents a department, and the edges represent direct managerial relationships between departments.
• The company consists of n departments, labeled from 0 to n - 1.
• You are given n - 1 manager-subordinate relationships in the form of an array edges, where each element [a_i, b_i] indicates that department a_i is the direct manager of department b_i, or vice versa. This structure forms a tree (i.e., there are no circular managerial chains).
Task:
Your goal is to choose the most efficient department to serve as the central management hub for the entire company. The height of the hierarchy represents the longest managerial chain (in terms of reporting lines) from the central management department to the most distant department.
In a well-designed organizational structure, you want the central management department to be as close as possible to all other departments, minimizing the distance (in terms of reporting lines) to any department.
Objective:
You need to identify which departments can serve as the best central management hubs. These are the departments that, when selected as the central hub, result in the shortest longest reporting chain to the most distant department. This ensures that communication and decision-making processes are as efficient as possible.
Output:
Return a list of all the departments (node labels) that can act as the best central management hubs for the company. The answer can be returned in any order.
Input: n = 4, edges = [[1,0],[1,2],[1,3]]
Output: [1]
Input: n = 6, edges = [[3,0],[3,1],[3,2],[3,4],[5,4]]
Output: [3,4]
O(Number of nodes)

Question 2:

Imagine you're working with an e-commerce platform that tracks customer interactions with products. The platform collects data about which products users view, add to their cart, and purchase. As part of the product recommendation system, you need to identify the most popular products, i.e., the products that are viewed the most often.
Problem Statement:
Given an array nums representing the product IDs that customers have viewed, and an integer k, you need to identify the k most frequent product IDs. These products are the ones that are viewed most often, and they will be recommended to users.
Example:
• Product Views (nums): [1, 2, 3, 1, 1, 4, 2, 2, 5, 6, 2]
• k: 3
Output: [2, 1, 3]

Cleared this round as well, within another 10 minutes got call for appearing in another round.

Round 3: LLD Round

Interviewer asked my current organisation project, and said you have to design something like you might have seen in your current organsation.
So design LLD of Notification Service.
The notification will be sent to multiple platform such as SMS,WHATSAPP,PUSH notification.

Discussed the approach,Design patterns I will use and he asked some follow up questions.
Then wrote the high level classes for them , drew a diagram as well.
He asked some question on tracking activity of the user on the notification.

Verdict: Selected

Round 4: HM Round

He asked me some general questions, then asked me some questions related to microservices as I worked too much on them, then asked me why you created this microservice as this can be accomodated in single monolith.

I answered this question but he was not satisfied, as after a point of time he asked me who took the decision of creating this microservice, I told upper management, So he asked me
Tell me about a time when you did not agree with your upper management.
I told him a scenario.

Then he asked me one damn wierd questions on which I went kind of blank.

Design a water bottle.

I was like what dude what you are asking.
Then he gave me some hint, shape, color,material, way of opening closing, capacity these will be there in water bottle, how you will use encapsulation,inheritance in the classes you wrote.

In the end he was not completely satisfied.
Verdict(Self): Lean Hire

Hey LeetCoders! 👋

Following up on my recent Goldman Sachs Java Interview Questions (Link), I wanted to share another experience — this time with Adobe, where I interviewed for a Full Stack Java Developer role.

These are the questions I was asked during the technical rounds:

1. How is a thread created in Java?
2. How would you maintain concurrency when multiple requests come to your server?
3. How do you convert a HashMap into a ConcurrentHashMap?
4. What is the difference between class lock, object lock, and method lock in Java?

🧠 Know the answers?
Drop your thoughts and explanations in the comments section below. Let’s create a learning hub for everyone preparing for similar interviews. 💬

🚀 Also check out:
🔗 My Goldman Sachs Java Interview Experience

Happy coding and good luck with your interviews! 🙌
#Adobe #Java #Concurrency #FullStack #InterviewExperience #LeetCode

Let me know if you'd like help cross-posting this to LinkedIn or Medium too!

Technical Interview – Problem Solving, C++, OS
Interviewer Experience: ~19 years
Status: Rejected

Recruiter told : There is no detailed feedback as of now. If there is anything that can be shared, we will make sure to do so as soon as possible.

I don't know what does it mean. It's been more than two days.

C++ Question (15–20 mins)
The interviewer started with a deep dive into C++ smart pointers, specifically focusing on std::unique_ptr and move semantics.

He shared a code snippet and asked:

Will this code compile and work? If not, why? How would you fix it?

We discussed various things around ownership transfer,and how std::move interacts with unique_ptr.

Also, about modern C++ memory management.
Pass by refernece and pass by value.

Coding Question: Cousins in Binary Tree - Given a binary tree and a target node ptr, return all its cousins

I first proposed a 2-pass BFS approach. We then briefly discussed how it could be optimized to a 1-pass BFS by tracking parent and level simultaneously. There was a bit of miscommunication while I was trying to explain the 1-pass version, so the interviewer asked me to code the 2-pass approach, which I completed.

Link to Round 1 Interview Experience : https://leetcode.com/discuss/post/6722120/adobe-computer-scientist-1-round-1-inter-xjos/

Recruiter didn't share any feedback

Adobe Interview Experience – C++ (CS-1 Role)

YOE: 3

The interviewer started with a quick introduction and then moved on to C++ fundamentals. We discussed:

• vector vs list vs queue: use-cases, internal structure, and time complexity of operations.
• ordered_map (i.e., std::map) vs unordered_map, including internal implementations (Red-Black Tree vs Hash Table).
• What is a VTable and how virtual functions are implemented in C++.
• Implement a Singleton class in C++, covering thread-safety. Asked for some real use case , I tolder Logger.

Coding : I believe it was a hard-level problem. Problem: Minimize the Maximum Beauty of K Subarrays Given an array, divide it into k contiguous non-empty subarrays. The beauty of a subarray is defined as: beauty = max_element × length For each split, compute the maximum beauty among the k subarrays. Among all such splits, return the minimum possible maxium beauty.

I told the approach and coded but couldn't make it work, but he was satisfied with the thought process and approach.

Toward the end, the interviewer switched to OS questions, such as:

What is virtual memory? Paging and Segmentation

Questions based on code snippets: I was asked to analyze a few C++ code pasted in the editor .

Recruiter reached out with positive feedback.

Previous Experience: SDE 2 at FAANG (7 YOE)

Recruiter reached out after an Employee Referral. Process took 2 weeks

Round 1 - Coding

Question 1 - https://leetcode.com/problems/nested-list-weight-sum/

You are given a nested list of integers nestedList. Each element is either an integer or a list whose elements may also be integers or other lists.

The depth of an integer is the number of lists that it is inside of. For example, the nested list [100, [20,2], [[3], 2],1] has each integer's value set to its depth.

Return the sum of each integer in nestedList multiplied by its depth. Example 1: Input: nestedList = [[1,1],2, [1,1]] Output: 10

Explanation: Four 1's at depth 2, one 2 at depth 1. 1x2 + 1x2 + 2x1 + 1x2 + 1x2 = 10.

Example 2: Input: nestedList = [1, [4, [6]]] Output: 27 Explanation: 1x1 + 4x2 + 6x3 = 27

Question 2 - https://leetcode.com/problems/valid-palindrome-ii/description/

Round 2 - LLD

Was asked LLD for chess. Had discussions around classes, data members and member functions. Was asked to write the classes and method signatures.

Round 3 - HM Round

• He asked about my most critical project.
• Few C++ thoery questions - null pointer, dangling pointer, void pointer, memory management and smart pointers.
• Few C++ output based questions, like what is the output of this code snippet.
• Few C++ debugging questions - There was a bug in pasted snippet. I had to find out the bug
• Coding question - https://leetcode.com/problems/string-compression/description/
• Coding question - https://leetcode.com/problems/compare-version-numbers/description/ (Same question but the number of dots were fixed to 2)
• In Coding questions, the focus was on running the code on complier. Had to write main functions, input, output, imports everything.

Round 4 - Director Round

• Started with one behavioral questions around my past projects. (10 mins)
• After that - https://leetcode.com/problems/lru-cache/description/ (They didn't asked me directly this question. Gave a real world scenario and had discussion on what algo to use to solve this and then asked me to write the class and methods. Expectation was working code in C++)

Recruiter confirmed next day that I have cleared the interviews.

Adobe SDE-3 Frontend Interview Experience (4 YOE, 75 LPA Expected)

Total Rounds: 3 (All in one go, then Manager Round later) Mode: Entirely virtual – no calls, no role clarity, minimal communication

Recruiter Outreach: Got a cold email from Adobe’s recruiter asking for basic info – CTC, current location, experience, etc., and if I was available that Friday. No intro call, no JD shared, no explanation of the process or expectations.

My Profile:

4+ Years of Experience

Frontend-heavy background

Expected CTC: 75 LPA Previous CTC: 60 LPA

🧪 Technical Rounds (Back-to-back, same day): 1️⃣ HTML/CSS Round (8 mins) Had to recreate a simple figure using plain HTML & CSS. Straightforward if you’ve done basic layouting – wrapped this up quickly.

2️⃣ DSA Round – Leetcode Medium Build a custom Stack with support for deletion at any position in O(1) time. ➡️ Required a Doubly Linked List Solid DSA knowledge and quick debugging helped me solve it within time.

Managerial Round (Delayed & Rescheduled multiple times) Took place several days later due to availability issues.

Discussed my resume briefly.

Ended with a fun logic puzzle about finding the top 3 fastest horses with limited race tracks – standard version of the problem.

Outcome

After clearing all rounds and waiting for updates – got a rejection. No explanation. No feedback. No follow-up call. Nothing.

📝 Final Thoughts:

• No JD or role clarity upfront.

• Rejection without reasoning, after completing everything, felt very disrespectful to the candidate’s time from such a big organization

Previous Experience: a MNC(2+ Years)

Recruited reached out directly.
Whole process took 3 weeks from the first interview to last.

1st Round: 1 coding question(given i/p arr of len n of all positve numbers. You need to check if you can convert it to a zero array. You are also given a 2d array of queries having elements as (i, j) representing range. This is the range where we will be reducing the numbers in the original array by 1. Return if a zero array could be formed) + question about amazon s3 and some ML questions(due to my ML background)
2nd round: 1 coding question (https://www.naukri.com/code360/problems/the-celebrity-problem_982769) + design a universal query maker.
3rd round: 1 coding question(https://leetcode.com/problems/subsets/description/)+ questions on resume + questions on past projects + creating an api signiture for a chat completion + give positive and negative test cases for the api.(This was the worst round I had)
4th round: 1 coding question( https://leetcode.com/problems/cousins-in-binary-tree/ but instead of a binary tree an N-ary tree was given)
questions about resume, past projects, rough HLD about a basic flask app, behavioral questions. (This was the best round I had)

Verdict: Recieved the call from the recruiter after a 2-3 days with an offer.

The interview process for the MTS-1 role consisted of four technical rounds, starting with an initial phone screening by HR. The HR representative explained the job profile, collected details on relevant experience, and scheduled the upcoming interviews.

Round 1 and Round 2 were technical interviews focused on data structures, algorithms, and core computer science concepts.

Round 3 was a managerial round, which involved both technical deep-dives and project-related discussions.

Round 4 was the Director's round and interviewer was a senior manager, where both technical knowledge and career motivations were evaluated.

Each round progressively increased in depth and complexity and each one was elimination round. After clearing all rounds, a final confirmation call was made by HR informing the selection for the role.

I had an interview with Adobe Noida for the role of Computer Scientist 1.

Round 1:
They asked me two questions:
1. https://leetcode.com/problems/spiral-matrix-ii/description/
2. https://leetcode.com/problems/implement-queue-using-stacks/description/

I solved both questions during the interview.

For the first question, the interviewer showed me the full spiral matrix and explained that this is what needs to be printed. I might have misunderstood, and instead of printing the matrix only , i generated the matrix and print the matrix. When I clarified this with him, he said it’s okay since the logic is the same.

For the second question, I initially gave a working solution, but he wasn’t satisfied as it wasn’t optimized. After that, I explained and implemented the optimized version. I even submitted it on LeetCode after the interview and it worked perfectly. It was a slightly different approach from the official solution, but it was optimized.

However, to my surprise, when I reached out to the recruiter, she mentioned the feedback said I didn’t solve the first question and that the second solution wasn’t optimized.

It’s disheartening when things like this happen. I don’t memorize solutions — even for popular problems. I focus on understanding and solving them myself. i have solved 1000+ quesiton on leeetcode still i missed to solved this second question. I don't know whether he was expecting me to solve immediately as it was famous question.

Adobe SDE-2(C++) Interview Experience

Recently, my resume was shortlisted for the SDE-2 role at Adobe. The interview process was smooth and well-organized. Within 24 hours I received verdict of prev round and next rounds timings.

There were 4 rounds->

• First two technical
• Third HM(Techno Managerial Round)
• Director Round

Round 1 (Technical Interview 1 hour)

1. Design a Logger class
2. What is a virtual function( VPtr, Vtable )?
3. What is unique_ptr?
4. How do you delete a copy constructor?

   There are two ways:

   1. By making the constructor private
   2. Using the delete keyword
5. Function Overloading (with a code example)
6. Operator Overloading (with a code example)
7. Create static and dynamic objects

Note: You are expected to write code for all the above questions.

Round 2 (Technical Interview 1 hour)

This round covered both conceptual and coding questions:

C++ Basics: Questions on fundamental C++ concepts

• diamond problem and virtual functions

Operating System Concepts: Paging, Mutex, and Semaphore

Coding Problems:

1. Mirror Tree
   https://leetcode.com/problems/invert-binary-tree/description/
2. Given an array, determine if it can be split into two separate arrays such that both have unique frequency distributions.
   I gave brute better both approaches and he wants me to code for better one.
3. KMP Algorithm for String Matching
   I gave brute and better both approaches and he was happy with my code as well.

I told about time and space complexities as well.

Hiring Manager (HM) Round 45 minutes

Rate yourself in C++

Justify your rating by giving relevant experience where you individually tackle any problem in your project and solved it too(P.S :This will make good impact).

Discussion on past experience and projects

She asked many questions from resume.

General questions about my approach to problem-solving

Advise: You must know everything you written in your resume.

Final Verdict: ❌ Rejected

Although I didn’t make it this time, the experience was insightful. I will focus on improving my C++ fundamentals and problem-solving skills for future opportunities🚀

I interviewed with Adobe for an MTS 2 JAVA role, coming from a Tier 3 college with 2+ years of experience. This was my first time interviewing with the company.

Round 1 (Problem Solving)

1. Tell whether at least two non-overlapping increasing subarrays of length k exist in an array.
2. Find the maximum length of a subarray such that it contains at least two non-overlapping increasing subarrays.

Round 2 (Problem Solving - Java Focus)

1. Implement a concurrent HashMap with get, put, and delete operations.
2. Given a tree where each node has left, right and parent pointers, implement lock() and unlock() methods for a node. The condition for locking a node was that all its ancestor nodes and all its descendant nodes must be unlocked.

Hi, this is my first post on any platform sharing my recent interview experience for an MTS 2 Frontend role at Adobe. I went through five rounds, and I'm thrilled to share that I was selected!

Round 1: Technical (Frontend & DS Algo)

Round 2: Technical (System Design & DS Algo)

Round 3: Hiring Manager (Behavioral & DS Algo)

Round 4: Senior Hiring Manager (Technical & Reasoning)

Round 5: Design & Architect

My Adobe Interview Experience

I recently went through the interview process at Adobe for a Senior Software Engineer role. The process consisted of several rounds, which I found to be quite comprehensive and technically challenging. I've slightly modified the questions due to NDA.

1. Coding Round

This round involved two coding problems:

1. The first problem was based on Reverse Nodes in k-Group, but with an interesting twist. The interviewer asked me to reverse nodes in alternating sets of size x and y. For example, I needed to reverse x nodes, then reverse y nodes, and so on.
2. The second problem involved building an auto-complete system based on frequency. Given product names and their frequencies, I had to return the top 3 product names that matched a given prefix, sorted by frequency. The interviewer suggested a similar problem to consider was Longest Common Suffix Queries, but the core problem was distinct.

2. Concurrency Round

Adobe has a dedicated round focused on multi-threading, which I found quite in-depth. I was tasked with creating a task scheduler that supports two main functionalities:

• Single execution: Tasks should be executed once at a given start time, or as soon as a thread becomes available.
• Periodic execution: Tasks needed to be executed repeatedly after a given delay, for a specified number of iterations.

Initially, I proposed a min-heap solution where worker threads would continuously poll for tasks. However, the interviewer specifically guided me to implement a solution using Java's wait(), notify(), and notifyAll() methods. We also had a detailed discussion about the usage and implications of volatile and synchronized keywords in Java.

3. System Design Round

The system design problem involved designing a comprehensive webinar platform. The platform needed to cater to three types of licenses: Silver, Gold, and Platinum. Each license tier would determine:

• The maximum number of participants allowed per individual webinar.
• The overall participant limit for all ongoing webinars a user could host simultaneously.

For instance, a Gold license would permit 100 participants per webinar and a total of 300 participants across all of a user's active webinars. Additionally, the platform had a global constraint of 60,000 active users that needed to be accounted for in the design.

4. Hiring Manager Round

This round was more technical than I had anticipated. The hiring manager presented a real-world problem their team was facing: scaling a legacy monolithic system. The system was originally built for a maximum of 100,000 users, but now they needed to scale it to 1 million users. The critical constraint was that the existing monolith couldn't be modified. I proposed an approach where we could deploy 10 instances of the monolith, with each handling 100,000 users. An intermediate microservice would be responsible for directing users to the correct instance using consistent hashing based on the user's email. To ensure uniqueness, the user ID from the monolith would be combined with the instance number.

My interview journey for the MTS 2 role at Adobe consisted of four comprehensive rounds.

Round 1 focused heavily on JavaScript. I was asked about object manipulation using destructuring, creating a polyfill for the Promise.all API, and handling event triggering during scroll and window resize, specifically discussing throttling techniques including naive and timestamp-based implementations.

Round 2 continued with a mix of DSA and JS. I had a problem on finding the Lowest Common Ancestor (LCA) of a Binary Tree. We also discussed Critical Rendering Path (CRP) and asynchronous operations. Further JavaScript concepts like the event loop, objects, call, apply, and bind were covered.

Round 3 began with a data structures question, specifically Level Order Traversal of a Binary Tree. The latter part of this round involved discussions about my past project experiences.

Round 4 was primarily focused on algorithms. I was tasked with reversing words in a long string, finding all direct and indirect flights from a given source to destination, and inverting a Binary Tree.

Adobe MTS 2 Interview Journey

I recently had the opportunity to interview for the MTS-2 role at Adobe in Noida. The process began after HR contacted me via Instahyre regarding an SDE position. Following an initial call, my first technical round was scheduled.

Technical Round 1

This round primarily focused on algorithmic problem-solving and included two challenging questions. The difficulty ranged from medium to hard on LeetCode. First, I was asked to find the longest continuous sequence in an array, a classic problem requiring a strong understanding of array manipulation and dynamic programming. Fortunately, I had encountered and solved this problem previously, which helped me complete it efficiently.

The second question was directly from LeetCode: "Apply Operations to Make Two Strings Equal." This problem tested my string manipulation skills and required an analytical approach to find an optimal solution. We spent about 10 minutes discussing various approaches before I wrote the code, which successfully passed the test cases presented by the interviewer.

Technical Round 2

The second technical round was again centered on Data Structures and Algorithms, with a particular emphasis on modified binary search. The interviewer presented an interesting twist: the mid-point in the binary search was generated by a random function. My task was to write code for numbers that would be searched appropriately with this modified binary search approach. The discussion also included follow-up questions aimed at identifying elements in an array that would be correctly found using this particular binary search method.

EM Round

The third technical round, led by an Engineering Manager, delved into system design and practical scenarios related to my prior job experience. The interviewer asked questions on operating systems to assess my understanding of core concepts and their real-world applications. We also discussed my current job responsibilities and past projects, which provided insight into my professional background. I was asked to elaborate on the system design of WhatsApp, which naturally led to discussions about message queues and scalability challenges. Furthermore, I shared my long-term career goals and explained why Adobe was an ideal fit for my aspirations.

Director Round

The final round was with a Director, starting with general discussions about job-related queries before moving into technical discussions on data structures and system architecture. The conversation began with designing an airline system, focusing on graph theory and complexity analysis. Additionally, we talked about caching mechanisms and on-call support, addressing practical challenges encountered in software development and maintenance. I also shared my perspectives on monitoring tools and request authentication mechanisms, demonstrating my understanding of industry best practices and emerging trends.

Verdict

I later received a call from HR confirming my selection for the MTS-2 role, which was a very satisfying outcome.

My interview process at Adobe consisted of three intense rounds.

Round 1 was focused on Data Structures and Algorithms, along with some system design concepts. I was asked to solve the Best Time to Buy and Sell Stock IV problem, which I managed to solve. Another DSA problem was Amount of Time for Binary Tree to Be Infected, for which I discussed both BFS and DFS approaches. Additionally, there was a discussion about the cons of Microservices over Monolith.

Round 2 was a mix of DSA, SQL, Java internals, and design patterns. I tackled questions on implementing a Thread-Safe LRU Cache and finding the Two-Sum in a sorted Doubly Linked List. I also had to demonstrate my SQL knowledge with a query involving various joins like LEFT OUTER JOIN, where I explained the logic, though I wasn't completely sure about the exact syntax. We discussed why Java doesn't allow two parent classes in inheritance and how to implement a thread-safe Singleton Design Pattern.

Round 3 was purely focused on system design. I was asked about the benefits, features, and USPs of Redis. The main challenge was the Low-Level Design (LLD) of Kafka as a message queue. I successfully discussed and implemented Publisher, Subscriber, and Topic classes using the Observer Design Pattern, but I struggled with implementing the offset logic for multiple subscribers of a topic.

Ultimately, the slight mess-up in Round 3 with the Kafka LLD offset logic led to my rejection. The bar at Adobe is indeed quite high these days.

I hold a B. Tech. degree from a reputable state-level college. Currently, I'm working as an SDET-2 at a prominent enterprise security & CDN product firm, with approximately 5 years and 4 months of experience.

During my notice period with other offers in hand, a recruiter from Adobe reached out to me for a white box testing role. I dedicated a few days to refreshing my Python skills, and the entire process took about three weeks. I requested the recruiter to expedite the process to help me make a decision regarding my existing offers.

Round 1

A Senior Computer Scientist conducted this round. The focus was primarily on the implementation aspects of both frontend and backend automation frameworks within my current projects. I was asked how I would implement such a framework from scratch using Python. Despite being out of touch with Python for over a year, I explained the latter part reasonably well, and the interviewer delved into fine details. Towards the end, I was presented with the Decode String problem. Having not seen it before, I devised a stack-based solution and coded it up. While a few edge cases failed, the interviewer was satisfied, and we quickly moved on to discussing test cases.

Round 2

Another Computer Scientist 2 led my second round. We began with relational databases, covering topics like DB indices and B-Trees. I provided satisfactory answers, although I hadn't extensively prepared in that area. Following this, I was given a medium-difficulty SQL query question, which I managed to solve after some thought. The round then shifted to a backend framework design problem for a book subscription and online reading service.

I started by outlining the necessary APIs and their payload structures. I explained how to handle various scenarios such as subscription management, integrating payment gateways, and implementing reading limits for non-subscribers. I then detailed my approach using class diagrams to create a backend test framework for the service. The interviewer asked several design-related questions, which I addressed comprehensively. I chose a Java-based design as I felt more comfortable with it in an interview setting compared to Python.

Both the first and second rounds took place on the same day. The recruiter contacted me the next day to schedule another technical round, emphasizing that each round was an eliminator.

Round 3

I believe another Senior Computer Scientist conducted this round. It started with Python-specific questions, where my answers were not entirely spot-on. He then inquired about creating CI/CD pipelines, test reporting metrics, and Jenkins. After that, I was asked to solve a problem involving reading lines from a large file using Python. I provided an answer but struggled with the Python syntax – I later realized I should have leveraged generators for better efficiency. Finally, he asked an easy binary tree question, which I solved using Java.

My performance in the Python section of Round 3 might not have been up to par, so the team took a week to evaluate me alongside other candidates. I proactively asked the recruiter to schedule another technical round if needed. After a week, the recruiter called to schedule the managerial round.

Round 4

This round involved standard managerial questions about my past projects and experiences. I was again asked an API framework design question for a service, which I answered verbally. I also used this opportunity to inquire about the role and other specifics. The call concluded about 20 minutes earlier than scheduled.

The team deliberated for two days on a hire/no-hire decision, noting my strength in Java-based frameworks but relative weakness in Python, which was a requirement for the role. Ultimately, the recruiter informed me that they would like to offer me the position.

I received the offer in October 2022 for the Computer Scientist 1 level in Bangalore. The total compensation package included a base salary of ~30.5L (including Employer PF + Gratuity), signing bonuses of 3L + 3L (second part after 12 months), an Annual Incentive Pay (AIP) of 15% of base (~4.2L), and stock options worth 75k USD over 4 years (~14.8L annually). The total first-year compensation was approximately ~55L, along with standard Adobe benefits. I had previously received offers from Visa, Fi.Money, and PingIdentity, all in the 35-37L ballpark.

I successfully negotiated once, getting the joining bonus and stocks bumped up. The base salary seemed a bit low, but I couldn't negotiate further on it, likely due to not having a similar competing offer that matched Adobe's total compensation.

I went through the initial screening process for the MTS-1 position at Adobe. My first technical round focused on problem-solving. I was presented with a classic grid-traversal problem. After successfully navigating that round, I made the decision to withdraw my application before proceeding to subsequent rounds, as I had already secured a position with another firm. Overall, it was a valuable experience.

I recently interviewed for a CS1 position at Adobe in Bangalore. My experience level is 4.5 years of experience, and I'm pleased to share that I received an offer after completing five rounds.

Round 1: Data Structures & Algorithms

• Rotting Oranges
• Binary Search Tree to Greater Sum Tree

Round 2: Data Structures & Algorithms

• Maximum Points You Can Obtain from Cards
• Longest Substring with At Most K Distinct Characters

Round 3: System Design

Round 4: Technical & Behavioral

Round 5: Behavioral with Director

I applied for a CS1 position at Adobe through an employee referral and received an interview call within two weeks. The entire process lasted about 4-5 weeks.

Round 1: Technical
This round focused on data structures and algorithms. I was asked to solve two LeetCode problems: Group Anagrams and Merge Intervals.

Round 2: Technical
Another technical round where I tackled the Median of Two Sorted Arrays problem.

Round 3: System Design
The focus here was on the Low-Level Design (LLD) of a client-side rate limiting library. The interviewer expected me to code a rate-limiting algorithm. I designed and explained a complete, thread-safe solution with a driver class demonstrating client ID-based rate limiting. We then discussed potential test cases for this code, including how to test threaded clients here, etc.

Round 4: Hiring Manager
This round started with extensive discussions about my previous work experience. The interviewer asked some quick questions on concurrency, binary semaphores, and mutexes.

Then, an open-ended question about linear search in an array was posed. I was asked to write an algorithm to search an element in an unsorted array, then modify the loop to be unbounded, and finally, address potential issues without using try/catch blocks. I explained that an unbounded loop would eventually lead to memory issues like stack overflow or integer overflow. I suggested increasing JVM memory or breaking the loop under certain conditions, but acknowledged that the linear search algorithm itself would eventually fail for some test cases. I attempted to use some things like bit manipulations or 'hacks' to control the infinite loop by decrementing the loop variable, but these weren't what the interviewer was looking for. I realized I wasn't clear on their expectations, which I believe was a turning point. Later, typical operational questions followed, such as how I handle conflict, what my manager would say about me, and how I react to not receiving credit for my work. I answered these confidently.

Verdict: Rejected!

I had an interview experience at Adobe for the MTS 2 position, holding 1 year and 8 months of experience. The interview process consisted of five rounds, covering various aspects of technical knowledge.

Round 1

This round focused on LeetCode Medium problems. I was given two specific problems: one was the Longest Substring Without Repeating Characters, and the other was Get the Maximum Score. There were also some standard C++ questions.

Round 2

In this round, the interviewer presented a problem involving a running stream of integers. The task was to fetch the rank of an integer, where rank means the count of elements less than the current element. The discussion heavily focused on the choice of data structure and the time complexity implications. I was asked to write the code from scratch.

Round 3

This round involved a system design question: I had to design a movie booking system, specifically outlining the classes I would use and their interactions. Following this, I was asked to solve a problem: given a file, print the last 'n' lines from it. There were also some questions on memory management.

Round 4

The fourth round again involved coding problems. I was asked to solve the Longest Palindromic Substring problem and the Buy Sell Stocks III problem.

Round 5

The final round started with questions on Operating Systems and the time complexity of various sorting algorithms. I then had to solve a problem where, given an array, I needed to find triplets that sum to a given target without sorting the array. The last coding question was about finding the Lowest Common Ancestor (LCA) of a Binary Search Tree (BST).

Ultimately, I was rejected. I struggled to answer the OS questions sufficiently in the last round, and made a stupid mistake by writing the code for the LCA of a general binary tree instead of specifically for a BST. Despite the outcome, the interviewers were very friendly and helpful throughout the process.

My interview journey for the MTS-2 role at Adobe in Noida was insightful. I hold a B.Tech in CS and have 3.5 years of experience as an SDE in a product-based company. The process consisted of four rounds.

Technical Round 1 (1 hr):
This round included questions based on Binary Search, with a specific constraint to use exception handling for certain cases. I also encountered a variation of the chocolate distribution problem. Additionally, the interviewer delved into C++ and OS concepts.

Technical Round 2 (1 hr):
During this round, I was asked to implement a function to copy a linked list with random pointers. I had to write multiple test cases and dry run them until the interviewer was satisfied with my explanation. A significant portion of this round also involved questions on C++ and OS.

Manager Round (40 mins):
I tackled the Balanced Parenthesis Problem and a puzzle involving 100 people in a circle, which is essentially the Josephus Problem, requiring me to write generalized code for it. This round also included more questions on OS.

Director Round (1 hr):
The final round involved some behavioral questions and a discussion about my projects. C++ and OS concepts were revisited. A notable question involved implementing a solution to safely share a variable between multiple threads, similar to the Reader-Writer problem, which I had to compile and run on an IDE.

Overall, my opinion is that Adobe generally asks easy to medium level coding problems, but their primary emphasis is on OOP concepts.

Round 1: Coding and Problem Solving

This round focused on my coding and algorithmic skills. The interviewer presented two problems:

1. First, I was asked to determine if there exists any contiguous subarray within a given array that sums to zero.
2. Following that, the second question built upon the first, requiring me to find the longest contiguous subarray within the array that sums to zero.

Round 2: Low Level Design

This round delved into my object-oriented design capabilities. I was tasked with designing a single entry-exit, single-floor parking lot, emphasizing OOP principles. A follow-up to this question introduced multi-threading, where I had to explain how parking slots would be assigned to multiple vehicles concurrently, testing my knowledge of concurrent data structures, synchronized keywords, and locks.

Round 3: Managerial Round

The managerial round started with a deep dive into my past projects, discussing their architecture, challenges, and my contributions. After the project discussion, I was presented with two more programming questions:

1. I needed to find the minimum number of jumps required to reach the end of an array, with a specific instruction to solve it using a greedy approach in O(n) time complexity.
2. The second problem involved finding the largest mountain subarray within a given array.

I found it important to maintain extreme calm and composure during this round, as it could become frustrating at times.

Round 4: Directoral Round

I recently completed my directorial round and am currently awaiting feedback. I will update this post once I receive the outcome.

Eligibility Criteria

• CGPA: 60% in 10th, 12th and 70% in UG & PG
• No Standing Arrears
• Pre-Final Year Students
• Degree: B.Tech & M.Tech (All Branches)

Round 1: Online Assessment (MCQ + CODING)

The online assessment was conducted on CoCubes, and it was mandatory to attempt all three tests:

• Cognitive: 20 Questions, 20 min
• Computer Fundamentals: 20 Questions, 20 min
• Coding: 2 Questions, 45 min
• Gamified Assessment: 3 Tasks, 23 min

After this round, out of 600 candidates, only 27 were shortlisted for the Product Profile Interview based on their assessment scores.

Round 2: Resume Review

Next, I was asked to share my resume. An Adobe talent partner reviewed the applications holistically. After a long wait of 12 days, I received an email informing me that I had been chosen for the final interview process.

Round 3: Video Interview

The video interview was conducted on MS Teams. It began with the classic "Tell Me About Yourself" question, for which I was well prepared. Following that, I was asked to discuss the major differences between Compile-Time Polymorphism and Run-Time Polymorphism, which I explained in great detail. I also faced some questions related to Operating Systems, specifically explaining the concept of Paging and Scheduling Algorithms.

Moving on to the programming part, I was asked to share my screen and code on my preferred IDE. The first problem was to convert an Infix Expression to Postfix Expression. The second question was to write the code to Reverse a given Linked List. I successfully solved both coding questions and proceeded further.

Next, I was asked to differentiate between Arrays & Linked Lists. After that, I was presented with a scenario: if I had to write the code for UNDO & REDO functionalities, which Data Structure would I use and why? I answered STACK, explaining it was due to the LIFO principle, and the interviewer seemed quite satisfied. He then moved on to questions about Non-Linear Data Structures like Binary Trees and Binary Search Trees. He also asked about the worst-case form of a BST, to which I replied Skewed Tree. The follow-up question was how to solve this problem, and I mentioned Tree Rotations.

Before the interview concluded, the interviewer asked if I had any questions for him, and then he signed off, wishing me luck.

Result

In just 1 day, I received my selection mail from Adobe for the role of Product Intern '22, and I was incredibly excited. It was an overall incredible process that turned out to be the best thing that happened to me.

First Round (~1Hr)

I was asked to code standard LeetCode questions with a bit of follow-ups and a code walkthrough. The problems included a LeetCode hard question similar to reversing nodes in a group of K, Course Schedule, and LRU Cache.

Second Round (~50Mins)

In this round, I was grilled on various C++ concepts. The primary focus was on Multi-threading (mutex, condition variable), Smart Pointers, lambda internals, and move semantics. I was also asked to code a multi-threaded program and answer a question based on a priority queue.

Third Round - Low Level Design (~1.5Hr)

This round was with the hiring manager. My task was to design a file explorer within an IDE. I was probed on various design patterns relevant to the problem and asked to write skeleton classes, ensuring the right abstraction. Behavioral questions were also included in this round.

Result & Final Round

I received positive feedback and was scheduled for a director round. The final round consisted of behavioral questions, a discussion on my current work and accomplishments, and some open-ended technical questions. I received an offer.

I recently interviewed for a C++ developer position at Adobe, which consisted of two distinct rounds.

In the first round, I was presented with a problem from LeetCode, specifically Jump Game IV. I attempted to solve this problem but was not able to fully complete it during the interview. Following that, I was given a small C++ question involving class instantiation and asked to identify if it would compile successfully. The code snippet was:

Class A{
}
int main(){
A obj;
A *obj=new A();
}

The second round featured another LeetCode problem, Open the Lock. Additionally, I was questioned about various C++ constructor types and tasked with implementing them for a custom String class.

I recently interviewed for the Computer Scientist II position at Adobe, located in Noida, India. My total experience at the time was over 10 years. The virtual interview was conducted using BlueJean.

Round 1 (1 hour)

The first round covered a mix of probability, puzzles, system design, and data structures:

1. We discussed a basic probability question involving biased dice.
2. A puzzle based on the Lucky 7 dice game was presented. I was asked about the strategy for this game.
3. I was asked to design the classic Snake game, with expectations for both high-level architecture and low-level class diagrams.
4. Finally, there was a question involving a BST tree where each node had additional information about the number of children in its left subtree. With limited memory, the interviewer wanted to know how to divide the tree into multiple pages (files/parts) given a maximum number of nodes per page. I implemented a solution using a getPage method, and you can refer to my code via the provided link.

Round 2 (1 hour)

The second round focused heavily on C++ internals, operating system concepts, and more design challenges:

1. We discussed move semantics in C++.
2. The difference between std::move and std::forward in C++ was questioned.
3. I was asked about methods pre-generated by the compiler in C++, specifically regarding C++11 and C++14 standards.
4. The interviewer inquired about the mutable keyword in C++.
5. Variadic templates in C++ were also a topic of discussion.
6. I had to explain the differences between static linking and dynamic linking.
7. The "copy on write" operating system concept was brought up.
8. A design problem involved creating a class with the restriction that only two objects of it could be instantiated, followed by a discussion on how to handle this in multi-threaded environments.
9. The final question required designing a class for managing meeting room intervals, including methods for adding/deleting intervals and getting the number of rooms, both generally and for a specific time range.

After completing Round 2, I did not receive any feedback call from HR. Based on this, I concluded that I was rejected for the position.

My interview experience for the MTS-2 role at Adobe Systems in Noida, India, was quite positive, with very friendly interviewers throughout. I was referred by a friend for this position.

Round 1 (40-50 min)

• The round started with a discussion about my projects.
• I was given a problem: Given an mn matrix and q queries (a b c d, where (a,b) are top-left and (c,d) are bottom-right coordinates), find the sum of elements in the rectangle for each query.
• Another problem involved a factory with 10 machines, all producing 100g balls except one defective machine producing 90g balls. I had to find the minimum weighings required to identify the defective machine using a weighing machine.

Round 2 (40-50 min)

• Again, the round began with a discussion about my projects.
• I was asked to find the next largest N-bit binary number with the same number of set bits as the original (e.g., 101 -> 110).
• A problem involved rearranging an array of 2n numbers (a1...an, b1...bn) into a1,b1,a2,b2,...an,bn without using extra space.
• There were conceptual questions: What is a constructor? What is a destructor? Follow-up: How to keep track of live objects of a class without using global variables.
• Another array manipulation problem: Given an array a1...an and an index i (1 <= i <= n), reverse the array around that index in-place (e.g., a1..a6, i=3 -> a4,a5,a6,a1,a2,a3).

Round 3 (40-50 min)

• This round also started with a discussion about my projects.
• I was tasked with a problem: Given a binary search tree, multiply all elements by -1, and then arrange the elements to again form a binary search tree.
• I was asked to list all sorting algorithm names and their time complexities. The follow-up question was: Given N sorted elements and M elements being added (M << N), which sorting algorithm would I prefer and why?

Round 4 (40-50 min)

• The first part of this round was a discussion about my projects.
• I was given a binary tree problem: Remove all nodes whose subtree has a sum less than K.
• Another problem involved finding the maximum number of gems I could collect from 'n' hills, where the i-th hill has 'ai' gems, and I couldn't collect from adjacent hills.
• A classic brain teaser: 100 ants are sprinkled on a rod, moving at 1 m/s and reversing direction on collision. Find the total time after which all ants will fall from the rod.

Round 5 (10-15 min)

• The final round was short and focused on a single problem: Write a mathematical expression to find the average of 4 integer numbers, avoiding integer overflow and only returning the integer part of the average (e.g., 3,3,4,3 -> 3).

Status and Background

I was working as an SDE-1 at Samsung Research Institute Delhi and graduated in 2018 with a B. Tech. in Computer Science from NIT Kurukshetra, India. My interview for the MTS-1 position at Adobe Systems in Noida, India took place on May 1, 2019.

1. Online Assessment Test (HackerRank)

The first stage was an online assessment on HackerRank. It comprised:

• 15 aptitude questions (15 mins)
• 5 input-output programs (15 mins)
• 2 programming questions (45 mins)

2. On-site Round 1 - Technical Round (1 Hour)

This round began with a discussion on my projects, including their architecture diagrams. Following that, I was asked several programming and theoretical questions:

• How to sort an array containing 0s and 1s in a single pass with O(n) time and O(1) space complexity.
• A problem to add 1 to a number represented as a singly linked list. Each node represented a digit, with the head pointing to the most significant digit. The challenge was to do this in a single traversal.
• I had to determine the number of different ways to form an N-bit binary number such that no two consecutive 1s occur. For N=3, examples like 101, 010, 000, 001, 100 are valid (output: 5), while 110, 111, 011 are invalid. I wrote a program, but the interviewer mentioned there's also a mathematical formula for it.
• I was asked about the Producer-Consumer problem, including writing pseudo-code. We also discussed semaphores, mutexes, critical sections, and potential deadlock scenarios involving wait() and signal() operations.

3. On-site Round 2 - Technical Round (1 Hour)

This round involved a deep dive into one of my projects. The interviewer skillfully transformed the project discussion into a problem and posed many follow-up questions. The core problem was:

How would you represent an image in memory, for example, as a matrix where each cell is a pixel? Considering a monochrome image represented by such a matrix, how would you programmatically draw a line segment on it? This meant filling the cells that the line segment would pass through. I had to account for both axis-aligned and non-aligned line segments and write a program for it. I initially assumed integer endpoints for the line segment and coded accordingly. The interviewer then challenged me on whether the code would work for float/double endpoints and we discussed generalizing the solution. We also covered boundary cases: what if the line segment's endpoints are outside the matrix, or if the segment is only partially inside? For a line segment not aligned to any axis and not bounded by the matrix, I was asked how to find the point where it first enters or exits the matrix.

4. On-site Round 3 - Technical Round (1 Hour)

This round covered a mix of C/C++ fundamentals, data structures, and algorithms:

• Questions on pointers, such as the validity of pointer differences and pass-by-reference in functions.
• What happens when an object of a class is declared and memory is allocated to it?
• Questions on object-oriented programming concepts like inheritance, polymorphism, and virtual functions.
• I was asked to implement a Hash Table with the following APIs: void insert(int key, int value), void delete(int key), int search(int key), and int getRecentElement(). The getRecentElement() API should return the key of the element most recently touched by either insert() or search(). All APIs were expected to operate in O(1) time, assuming no duplicate keys.
• A problem to search for an element in an array where each consecutive pair of elements differs by exactly 1 (i.e., |A[i] - A[i+1]| = 1 and |A[i-1] - A[i]| = 1).
• Given N threads and single instances of N different types of resources, how would I allocate resources to threads to avoid deadlock and minimize waiting time? This assumed threads could arrive at any time and request any number of resources, without knowing their arrival schedule.

5. On-site Round 4 - Director Round (1 Hour)

This was a broader discussion round:

• We discussed my projects.
• I was asked how to identify whether software I'm about to install on my PC is malware, providing methods for both pre-installation and post-installation.
• Following up on that, I was asked how to identify if the software had deleted or added files to my file system. I suggested comparing the file system's tree structure before and after installation.
• Building on the previous question, I was given two files containing lists of file names (representing states before and after installation) and asked to write pseudocode for a fileDifference(File A, File B) function to find the differences.
• A combinatorics problem: Given an N x N matrix, how many ways are there to reach the top-right corner from the bottom-left, moving only Up, Right, and Up-Right diagonally? I used a dynamic programming approach, but the interviewer prompted me to derive a mathematical formula. (I later learned this relates to Delannoy Numbers).

6. On-site Round 5 - Manager Round (1 Hour)

The final round was with a manager, focusing on problem-solving and logic:

• I had to write a program to convert a number into English word format, specifically using the Indian Currency system (e.g., 1,23,500 to “One Lakh Twenty Three Thousand Five Hundred”, 100 to “One Hundred”).
• Another task was to write a program to check if two given binary trees are mirror images of each other.