Phonepe Interviews

Before the final offer came, my journey wasn’t smooth. I had a few rejections, some interviews that went nowhere, and months of silence from companies I really wanted to join. I had kept my target list small: only a handful of tier-1 companies. For practice, I gave interviews at 2-3 service-based firms. Meanwhile, I was preparing consistently -

- Followed LeetCode 75 + 150,
- Completed a Udemy DSA course to build fundamentals,
- Read interview experiences from LeetCode + Google’s real-world interview blogs,
- And even got the chance to interview with Google and Amazon during this phase.

JPMC - Where It Started

I had applied to multiple positions on JPMC’s portal without any response. Then suddenly, in early June, I received the online assessment. The questions were easy - medium and I cleared all test cases, so I was hopeful… but HR never responded, even after multiple follow-ups. At that point, I thought it was over.

PhonePe - The Unexpected Turn

I initially applied for the SDE role, but it didn’t move. So I applied for the PSE role, and surprisingly got a call the very next day.

Round 1 - Technical
- One LeetCode medium coding question → solved
- Linux scripting + commands
- SQL queries (missed one complex one initially, but solved it after the interviewer reframed it)

Round 2 - Hiring Manager
- Deep dive into my TCS project work
- A few SQL queries again

End of August, I got the offer. The package wasn’t high, but PhonePe > TCS on my resume, and I wanted to move. I accepted and resigned. At this point, I had mentally closed the chapter on all tier-1 hopes.

Then… JPMC Called Back

After months of silence, in mid-September, I suddenly received a call:
“We’re moving forward with your application.” I honestly didn’t believe something would come out of it. But then the Super Day was scheduled - 3 rounds back-to-back (45 mins each + 15 min break).

Round 1 - Technical Debugging + Coding
- Given ~100 lines of code and asked to find bugs
- Apart from all technical bugs, I found a logical bug too, which the interviewer appreciated
- Then a sorting + pattern-maintenance problem with numeric array - completed ~80% before time ran out(nearly 1 hour long)

Round 2 — System Design
A very standard, commonly-practiced system design question. (Everyone who has done SD prep has seen this one.)

Round 3 — Techno-Managerial
More technical than managerial - framework choices, scenario-based problem solving, and reasoning.

After 1–2 days, I got the call:
I cleared the Super Day.

Then came team matches - I had two. Matched with the second team.

And finally… the offer rolled out.

My interview journey at PhonePe for the Software Engineer role, with 1-3 years of experience, began after I applied through Instahyre.

Round 0 – Online Assessment

The initial stage was an online coding assessment conducted on CodeSignal. This round comprised four Data Structures and Algorithms questions, ranging from medium to hard difficulty levels.

Round 1 – Coding Round

Approximately one week after completing the OA, I received an email from HR to schedule the first coding round. I requested to postpone it by two weeks as I was out of town. In this round, I was presented with the following problem:

• Min Stack

We thoroughly discussed multiple approaches to solve this problem. Due to time constraints, this was the only question covered in this round.

Round 2 – Coding Round

The very next day, I received a call from HR and scheduled the second coding round for the same week. This round involved two specific questions:

1. Accounts Merge
2. A variation of the Meeting Rooms problem, specifically focused on finding the minimum number of meeting rooms required.

Round 3 – HM Round

Again, HR contacted me the day after the second coding round, and I scheduled the Hiring Manager round for the following week. This round primarily involved a discussion about my resume and my work experience at my current company. I was asked technical and architectural questions related to my current projects. Additionally, the interviewer posed a few behavioral and situational questions to understand my approach to various workplace scenarios.

Verdict: I was selected for the position.

My interview journey at PhonePe involved several intense rounds, primarily focusing on Data Structures & Algorithms (DSA) and System Design.

DSA Rounds

The DSA rounds were comprehensive, covering a wide array of topics. I was presented with problems such as "Burning Tree" and "Shortest Path to Get All Keys". I also tackled LeetCode 1235: Maximum Profit in Job Scheduling and 1657: Determine if Two Strings Are Close. During Matchsticks to Square, I initially struggled due to misinterpreting the input format, but once I understood, my approach was satisfactory to the interviewer. For Kth Smallest Element in a Sorted Matrix, I started with a min-heap solution, and after a hint, optimized it using binary search.

Other notable DSA problems included Burst Balloons (a hard DP problem) and The Celebrity Problem. In a subsequent PS/DS round, I was asked to solve Remove K Digits and a problem from Codeforces. I also faced a custom question: "Find the minimum number of swaps to make a string palindrome", for which I used a greedy approach. Another question involved finding "the sum of all nodes at the deepest level in a binary tree".

I encountered a complex problem on maximizing total damage with spell casting, where I had to consider sliding window or dynamic programming, and removing the most stones in the same row or column, which required graph theory and Union-Find. Classic problems like Trapping Rain Water, Left View of a Binary Tree, and finding nodes at 'k' distance from a target node were also asked. I implemented a Min Stack and was tasked with printing the level order traversal of a tree with alternate direction changes, which I solved easily. Another monotonic stack problem was given, and despite modifications, I managed to solve the first two versions.

Additional problems included First Missing Positive, Longest Increasing Path in a Matrix, and an O(log n) optimization challenge for the Complete Binary Tree Inserter (which I could only solve in O(n)). Queue Reconstruction by Height, House Robber III, Kth Ancestor of a Tree Node, and Bricks Falling When Hit were also part of the technical assessments. I was also asked a question related to the Convex Hull Algorithm and how to generate random points within a circle with equal probability distribution.

System Design Rounds

The System Design rounds were equally challenging. I was asked to design a Search Autocomplete feature, where I successfully implemented a Trie-based solution and discussed scalability and optimizations. One of the final rounds was to design Instagram, where I focused on posting, immediate visibility, feed generation, and likes/comments. I covered NFRs, scale, HLD, and trade-offs, though I felt I could have elaborated more on the likes/comments service. I also had to design a service for delayed triggers, a generic online game (like Ludo/Chess) where I had to drive the discussion from FNRs to tech stack and matching logic, and Payment Gateway Integration. Another iteration of Design Instagram highlighted the interviewer's disagreement with a message queue for feed publishing/fetching at massive scale. Lastly, I designed Twitter, covering tweet posting, follow-following, timelines, and profiles, emphasizing LLD and HLD aspects.

I had a machine coding round at PhonePe, which focused on building an 'App Version Management System'. The problem statement detailed the need to manage different app versions, handling fresh installs and updates, similar to how smartphone apps evolve. The core idea was to design a system where app owners could directly interact with target devices, bypassing traditional marketplaces like Playstore or AppStore. Key functionalities included uploading new versions, creating update patches between versions, releasing versions with strategies like beta or percentage rollouts, and checking for app version compatibility and updates. I also had to implement functions to execute install or update tasks. The expectation was to have everything in memory, focusing on OOPs/OOD principles, code readability, and testability.

I recently went through the PhonePe hiring process (OA + 3 rounds) and wanted to share my experience. The OA had 4 questions, 2 from DP, 1 from Trees, and 1 unique problem. The last question required counting numbers greater than B formed from a string. I solved 2 fully and partially solved the others, which led to my shortlisting. In the first technical round, I tackled two LeetCode problems: Burning Tree and Shortest Path to Get All Keys, writing complete code and dry runs. The second round was more ad-hoc, and I made it to the third round. The third round had two problems: Codeforces 2028C and a custom problem involving maximizing a sum with constraints. I suggested bitmask DP for small n and wrote the solution. The feedback was positive, and I got shortlisted in the end.

Role: SDE-1
Platform: Onsite/Online (Coding Environment Provided)
Round Type: Machine Coding (Duration: 90–120 mins)
Language Used: Java

💼 Problem Statement:

Design a Fitness Class Booking System with the following features:

👥 User Management:

• Users can register/login with tier-based packages:
  • Platinum → 10 classes
  • Gold → 5 classes
  • Silver → 3 classes

📅 Class Management (Admin Functionality):

• Create, schedule, and cancel fitness classes (e.g., Yoga, Gym)
• Each class has a capacity and scheduled time

📌 Booking Functionality:

• Users can book classes based on available slots and their tier limit
• If the class is full → Add to waitlist
• On cancellation, the first waitlisted user gets the slot

❌ Cancellation:

• Users can cancel up to 30 minutes before the class
• Cancelled bookings restore user’s quota

🔄 Concurrency:

• Multiple users trying to book at the same time → Thread-safe booking logic was expected

🧠 My Approach:

• Used OOP Design: Classes like User, FitnessClass, BookingService, etc.
• Followed Separation of Concerns using service classes
• Stored everything in-memory using HashMaps/List (no DB required)
• Used synchronized blocks for thread safety
• Hardcoded the main flow (register, create class, book, cancel, etc.) for quick testing

✅ Features Implemented:

• User registration with package tiers
• Admin can schedule classes
• Class booking with capacity checks
• Waitlist promotion after cancellation
• Tier-wise booking limits
• Concurrency-safe booking

Feel free to ask if anyone wants the code or detailed breakdown. Hope this helps for your upcoming rounds! ✌️

YOE:4.6

Round1:
MachineCoding round

nextday: Explanation of code written in Machine coding round

verdict: positive

Round2:

PS/DS

1. https://leetcode.com/problems/maximum-total-damage-with-spell-casting/description/
2. https://leetcode.com/problems/most-stones-removed-with-same-row-or-column/description/

verdict: postive

Round 3:
System design

Instagram news feed

verdict:Rejected

DSA round 1

Q1. Given an array of integers arr and a integer k, you have to remove exactly k elements either from the start or the end or a mix of both.

Return maximum sum of integers that can be removed

Example 1:
Input:

	arr = [1,2,3,4,5]
	k = 2
	

Example 2:
Input:

	arr = [5,1,2,3,4]
	k = 2
	

Q2. You are given a string s. You can add any character at any place in the string. Output the minimum number of characters that have to be added to string to make it a palindrome

Example 1:
Input:

	s = "abcd"
	

Example 2:
Input:

	s = "abac"
	

Example 3:
Input:

	s = "abdab"
	

DSA round 2

Q1. Given a string s, and integer k, you have to perform this operation as many times as possible:

Remove k consecutive characters from s which are same, and concatenate the left and right part.

Return the string after the above operation can no longer be performed.

Example 1:
Input:

	s = "abdc"
	k = 2
	

Example 2:
Input:

	s = "abbadc"
	k = 2
	

Example 3:
Input:

	s = "pqqqqppeeeeepppqq"
	k = 5
	

Q2. You are given a undirected graph G in form of adjacency matrix of size n x n, containing n nodes, indexed from 0 to n-1. You are also given a list of infected nodes infected.

A healthy node is infected if it connected to the infected node. At time t=0 the nodes in infected are the only nodes that are infected. But after time, more and more nodes get infected, as the infection propagates. Let's say at t=infinite, there are MAX_INFECTED nodes.

Now you can disinfect one and only one, at the start at t=0. Your task is to find a node n, disinfecting which will minimise the MAX_INFECTED nodes count. Print that node's index. If more that one node can cause minimum MAX_INFECTED, return the node with smallest index.

Example 1:
Input:

	G = [[1, 1, 0], [1, 1, 0], [0, 0, 1]]
	infected = [0, 2]
	

Example 2:
Input:

	G = [[1, 1, 0], [1, 1, 0], [0, 0, 1]]
	infected = [1, 2]
	

Example 3:
Input:

	G = [[1, 1, 0], [1, 1, 0], [0, 0, 1]]
	infected = [0, 1]
	

Explanation

3 nodes are there. node 0 and node 1 are connected, and node 2 is not connected to anything.

• Example 1: removing infection from 0 will cause MAX_INFECTED = 1, which is minimum
• Example 2: removing infection from 1 will cause MAX_INFECTED = 1, which is minimum
• Example 3: removing infection from either 0 or 1 will cause MAX_INFECTED = 2, but 0 has smaller index, so res=0

Managerial round

• Intro
• Why changing company
• Why phonepe
• Current company project discussion in detail
• What and why in current company project
• Design blinkit. High variation in number of users of users and number of dark store per square area. (And a lot of roasting here)

Result

Rejected.

PhonePe is conducting its annual hackathon and wants to create an interactive platform to host the event. It will be a 2 day event and contestants need to maximize their score over this period by either solving more problems or more difficult problems. Design the backend to support this platform.

**Mandatory Functionalities ** The system should have the capability to add problems to the Questions Library. Contestants should be able to register themselves with their name and their department name. A problem should have attributes like description, tag, difficulty level (easy, medium, hard), score. Contestants should be able to filter problems based on difficulty level or tags and sort them based on score (design should be extensible to other attributes ) A contestant should be able to solve a problem as well as get the list of problems solved by him/her. A contestant should be able to see the number of users that have solved a given problem and average time taken to solve that problem. Scoring strategy for a problem could simply be to award the score assigned for the problem or could be something different like a combination of score and time. Return the current leader of the contest Users should be able to get curations like Top 10 most liked problems of a certain tag. Extension Problem

On solving a problem, users should get a recommendation of top 5 problems based on relevance The recommendation strategy for problems could simply be similar tags or extensible to include other factors like number of users who have solved a particular problem or a combination of factors ( Design should be extensible). Capabilities

Below are various functions that should be supported with necessary parameters passed. These are just suggestions, the signatures can be altered as long as the functionalities are provided.

Registering a problem - addProblem() Registering a user - addUser() Fetch the list of problems - fetchProblems() - Should take as input filtering and sorting criterias and return all matching problems in the right order. The result should contain problem attributes like name, tag, difficulty level, score etc. Should also display number of users who have solved a problem and the average time taken for that problem Solve a problem - solve() - Exposed to a contestant to mark a problem as solved [ For the extension problem, this function should return next 5 recommended problems ] Fetch solved problems for a user - fetchSolvedProblems() - Fetch the list of solved problems for a user Get leader - getLeader() - Returns the name and department of the leader Get top n liked problems of a certain tag - getTopNProblems()

Hi All,

I received phonepe interview after applying on their career site. Recruiter sent me a machine coding problem and I had 1 day to submit it.

Machine Coding Question Design and implement a battleship game to be played between two players until one comes out as the winner. A player wins When all the ships of his opponent has been destroyed. Missiles are fired randomly and they should not be fired at the same location twice.

Round 1 Solution discussion of machine coding round. Added functionality that there should be two modes automatic fire and manual fire. Follow up question what if missile has a range and it can destroy all the blocks in its range in all 8 directions.

Round 2 1235 Maximum profit in job scheduling

1657 Determine if two strings are close

solved first one completely and gave approach for the second one didn't have time to code second.

Round 3 System Design: Design a service which registers triggers with delays, from other services and sends them response of the trigger after the time is completed for that trigger.

Verdict: Rejected. Felt like system design interviewer was not nice at all, had lot of attitude. He started grilling me on API request and response body. Didn't even let me have a flow at any point in the interview.

I recently interviewed with PhonePe for their SDE II role. Here's how it went.

First Round – Machine Coding Round This round was scheduled by the recruiter via zoom, where I was provided a link to a coding pad with the question. I had to submit the solution via email within 1.5 hours. I was allowed to solve this offline on my editor. It was a typical object-oriented design question. Though I don’t have the exact question now, many similar problems are shared on GitHub by past candidates. I was expected to implement 6–7 functions. You are not expected to complete everything so do it sequentially. I managed to complete all 7 functions, even though there were mistakes in the last one.

The focus was clearly on implementing a working solution in a short timeframe while ensuring modularity, scalability, and OOPs.

The very next day, the recruiter reached out to schedule a follow-up with engineers to discuss the code.

Second Round – Code Review & Deep Dive This was essentially a code walkthrough with two engineers from the team. Using the example they'd provided, I explained the flow and logic across each function.

They asked questions around concurrency and thread safety. My code was not thread safe and I told them in the interest of velocity I have written this, but ya i would introduce locks otherwise. They also asked some questions around different scenarios and how I would implement that.

It felt more like a peer review than a grilling interview, and overall, the round went smoothly.

Third Round – DSA and Problem Solving I was asked two LeetCode-style questions:

Matchsticks to Square: I initially misread the example due to confusion around the input format, which cost me a bit of time. But once I understood the problem, I discussed my approach and time complexity. The interviewer seemed satisfied and we moved to the next question.

Kth Smallest Element in a Sorted Matrix: I discussed a bunch of solutions and decided that using a min heap would be the best solution. He hinted it could be optimized further. I could not think of a way, so he mentioned binary search, and gave me another hint around it, and then I could solve it.

This round went really well in my opinion. I prefer interviews like this, where the focus is on how you think, how you explore edge cases, build up your reasoning, and adapt with hints. I’d take this kind of discussion any day over an interview where he gets hung up on remembering syntax or using APIs/methods verbatim. This felt collaborative, not combative hahaha, and that makes a world of difference.

Fourth Round – System Design The final round was a system design round to design Instagram. After discussing scope and expectations, we narrowed it down to four main features:

1. Posting content
2. Seeing one's own posts immediately after posting.
3. Feed generation
4. Likes and Comments

I discussed the qualities of the system (non-functional) and set my scale expectations well. From there, I outlined a high-level design and then looked into each of the services. While I covered the core architecture and trade-offs, I realized later that I didn’t allocate enough time to flesh out the likes/comments service fully. I hand-waved a bit near the end as I was short on time.

Result: Rejected on the last round 😞

Outcome & Reflections The process was progressive—I would proceed to the next round only if I cleared the previous. Recruiter mentioned that the team had a meeting to discuss my case, since only the last round was not up to the mark. Ultimately, the feedback on the final round wasn’t strong enough to move ahead.

In hindsight, I could’ve paced the discussion better and spent more time detailing data flow and scaling for likes/comments. I did check with the interviewer throughout to ensure the direction was right—but I missed some depth where it mattered. That said, the round still felt constructive and insightful, had time allowed, I know I could’ve taken it further.

Problem Statement Design and implement an advertisement engine. The system manages Advertisers, Ads, and User Ad requests. It must select and serve the most relevant Ad based on matching criteria, bidding amount, and system constraints while respecting budget constraints.

Different terminology:

Advertiser: An entity that publishes advertisements on the platform. Advertisers also keep their advertisement budget on the platform. User: An entity that uses the platform and consumes advertisements with other use cases on the platform. Bid Amount: Bid amount is the amount earned per advertisement by the platform from the advertiser. If the Bid amount is higher, then there are more chances of the advertisement getting served to the customer. Advertisement Campaign: The Advertiser creates advertisement campaigns on the platform. The advertiser provides info such as URL, content_type (image, video), and bid per Ad. matching, and targeting attributes such as age, interest, and gender, etc., with the campaign. System Constraints: The Platform will have some constraints and will be checked against all the advertisements before matching. Advertisement Matching: The platform provides advertisements, comparing user details and attributes with all the active campaign requirements. It also compares the bid amount with the existing budget and checks for all the given system constraints for advertisement serving to find the best-matching advertisement. Once an advertisement is served, the budget is reduced from the advertiser’s account. An advertisement is served only when the bid amount is less than the existing budget.

Requirements P0 Provide interfaces to add advertisers and their budgets. add_advertiser(name) add_budget(budget) Provide interfaces to add users and their attributes to the system, such as date of birth, interests, gender, etc. add_user(user, dob, gender) add_attribbute(attributes) Provide an interface to create an advertisement campaign for the advertiser. create_campaign(advertiser_id, bid_amount, url, type, age, city, constraints) Provide an interface to request best best-matching advertisement from the system. This should return a single advertisement per API call. match_advertisement(user_id, city) All the above api syntaxes are symbolic. Please design and write the method names, method parameters as per the best coding practices.

System Constraints:

A user shouldn’t see the same advertisement if he has seen it in the last 10 fetch instances. At the global level, don’t serve the same advertisement if it has been served 5 times in the last 1 minute. match_advertisement can return null if all the advertisements after the criteria matching fail the system constraints validation for a user.

Requirements P1 Handle concurrency cases for different scenarios. Handling if more system constraints keep getting added to fetch advertisements.

Things to keep in mind You are only allowed to use in-memory data structures You are NOT allowed to use any databases You are NOT required to have a full-fledged web service or APIs exposed You are required to showcase the working of the above concept. Just a main class that simulates the above operations is enough. Provide some valid test cases as well. Should you have any doubts, you are allowed to make appropriate assumptions, as long as you can explain them during the evaluation. You are allowed to code on your favorite IDEs as long as you paste the code back into the tool within the allotted time frame How you will be evaluated You are expected to write production-quality code while implementing the requirements. We look for the following:

Separation of concerns Abstractions Application of OO design principles Testability Code readability Language proficiency [execution time limit] 1 seconds (cpp)

[memory limit] 2g

Software Engineer profile - Backend
4+ years of experience


Online Round:
Initial Coding Assessment - 2 Coding questions. Leetcode Medium.


Virtual Interviews:
1st Round: Machine Coding Round
Duration: 1 to 1.5 hours
- Was asked to create a working code of Payment Gateway System.
With Multiple features related to it.
Reference: www.youtube.com/watch?v=shipSEFMzHs

2nd Round: System Design Round
Duration: 1 hour
- Was asked to design a Search Autocomplete feature.
Trie data structure needs to be used and I created a entire working solution. Discussed on its scalablility and handle different scenarios and discussion on optimisation of memory and time complexity.

3rd Round: Hiring Manager Round
Duration: 0.5 hour
- Mostly questions were on why are you looking for a change and other behavioural questions regarding agile and others.



Got Selected!

Hi guys so recently I had the opportunity to interview with PhonePe as I was already on my notice period in Inmobi-Glance and I was having an offer from ShareChat which I also had shared earlier.

I got this interview through a referral from a PhonePe employee.

So the interview initially consisted of 4 Rounds only for SDE (Android) role. And those were:

• DSA Round
• Android Platform Round
• Machine Coding Round (Android)
• HM Round

Let's go thorugh each and every round one by one-

DSA Round - In this round I was asked 2 DSA questions. The time duration of this round was 1 hour only and I had to solve both the questions in that time limit only.

The first question was from Graphs topic and I must say that I am not very strong in Graphs and I was not expecting any Graphs question but it was my first question.

Question was similar to : https://leetcode.com/problems/loud-and-rich/description/

Literally I took a lot of time to firstly understand the problem then came to an unoptimised approach to which interviewer was not that happy.

Then after 30 minutes he presented me another question.

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

I solved this problem optimially before the given time limit and the interviewer was happy with my solution.

I totally lost my hope for next round but luckily I got call from recruiter the next day for next round :)

Android Platform Round - This round mainly revolved around basic android topics like ViewModel and its working, Activities, Fragments, Jetpack Compose.

Interviewer mainly dig deeper on topics like Services and its usecases which I comfortably answered.

There was no question from his side which I was not able to answer correctly.

Got a call from recruiter that I had cleared this round as well. Scheduled my next round the same day.

Machine Coding Round - In this round I was given a problem to design a E-Commerce app and how will I be managing the data between different screens.

The data should also be synced with the backend servers.

SO I basically was given some 4-5 criterias or features to complete in 90 minutes with scalable and clean code.

I followed MVVM + Clean Architecture in Android for this round. Firstly I told my approach to the interviewer and discussed a bit on this part.

Then when we were on same ground I started coding and I did it really fast as I had to complete all the features in the given time limit.

I did exceptionally well in this round that interviewer even praised me at last.

Then I got a call that I am eligible for HM Round. It was then scheduled for the next day.

HM Round - In this round the Hiring Manager discussed about my experience at Inmobi-Glance and I told whatever I had done in my 1.5 years of FTE at Inmobi-Glance.

Then he passed me an open ended question to design a map app and I had to tell him my approach in such a way that it is optimal and can be transformed into a market ready app with that approach.

We discussed a lot and then he asked some really tough behavioural questions to me which I answered confidently.

I felt this round as the most difficult one.

Unexpected happened : I was celebrating my farewell at my office (Inmobi-Glance) and I was pretty confident to get the offer that dat on May 30. Then HR called me and told me that there is a good and a bad news for me. I was shocked to hear this.

He told me that the collective feedback is mostly positive and they can consider me for an offer but I had to go through a Bar-Raiser Round due to my average performance in DSA Round

I literally was weeping from inside and multiple thoughts were running in my mind like: "May be they have found someone else that's why to reject me taking another round" etc etc.

But still I somehow managed myself and I agreed to his request.

The fact was that I also did not have any laptop to prepare for this round as I had submitted my mac back to my organization (Inmobi-Glance).

I borrowed a laptop from my friend and logged in my leetcode account and started preparing from next day.

Bar-Raiser Round - In this round I was asked 2 questions. And this round I would say was the most easy round.

The first question was based on "Min-Heap" which I solved optimally.

The second question was based on some strings like some word and pattern problem. I solved this also optimally.

Then that evening I got a call from recruiter that I had successfully cleared this round as well.

They were ready to give me an offer. And after 2-3 days I had my compensation call with my HM.

There we discussed my compensation.

Compensation details: https://leetcode.com/discuss/post/6817292/phonepe-offer-software-development-engin-e294/

Round 1 -

Machine coding pendency system

Round 2 -

Snake and ladder dsa

two pointer basic question

Round 3 -

Design shazam , hld

yoe: 2yrs applied online

previous round

was asked two DSA questions

1. https://leetcode.com/problems/burst-balloons/ (hard)
2. the celebrity problem (medium)

interviewer was nice and helped where ever they could. Wasn't able to solve any of them.

YOE: 2 years

Was given two hours to code this as an assignment. Question was held on codesignal, but you could use any IDE of your preference. You just need to mail the code. Interview was held 5 days later after submitting this assignment. You have to go through your code and explain your design.

Interviewer will judge you on the basis of your code quality, were you able to make it executable.

If you need C++ code for this just drop your email/linkedin/whatever. I can send you.

Verdict: was able to move to second round.

In interview

was asked to implement a method such that user shouldn't be able to book a class it has already booked a class in same timeslot. (was given 5 mins for that)

The task is to design a complaint resolution system to manage unsuccessful transactions. Key features include:

Customers can log issues for failed or pending transactions by providing details like issue type, subject, description, and email.

Customer Service Agents:

• Handle only one issue at a time.
• Can search issues by ID or customer email.
• Can resolve issues and receive new ones.

Admins:

• Can onboard agents with specific expertise (issue types).
• Can view the work history of agents.

System Requirements:

• Assign issues to free agents using an assignment strategy.
• Track and filter issues.
• Maintain resolution history.

Implementation requirements: createIssue(transactionId, issueType, subject, description, email) addAgent(agentEmail, agentName ,List) assignIssue(issueId) // -> Issue can be assigned to the agents based on different strategies. For now, assign to any one of the free agents. getIssues(filter) // -> issues against the provided filter, e.g.: getIssue({"email": "testUser2@test.com"}); updateIssue(issueId, status, resolution) resolveIssue(issueId, resolution) viewAgentsWorkHistory() // -> a list of issue which agents worked on

Recruiter reached out on 5th April through my application on Instahyre.

My details: Experience - 1 yr Current comp - 3 LPA

1st round - Coding assessment on CodeSignal for 70 mins It had 4 questions, 1 was easy, 2 were medium-hard, 1 was hard

I cleared the assessment after solving all questions.

Recruiter reached out a day after my assessment for 2nd round

2nd round - DSA and problem solving on Gmeet for one hour

There were two interviewers who asked leetcode medium questions, I dont exactly remember but they were related to arrays and trees.

Recruiter reached out on the same day for 2nd round and scheduled interview for 3 days later.

3rd round - DSA and problem solving on Gmeet for one hour Asked two questions-

1. Given a binary tree, find the list of elements which are at 'k' distance from the given target node.
2. Asked to implement stack and also to retrieve minimum element of the stack at any point of time.

Solved both the question in one hour.

Recruiter reached out for Hiring Manager round next day and scheduled after 4 days

4th round - Hiring Manager round Asked basic questions which are mentioned on my resume, and asked about the architecture and current project I'm working on in the current company. Later went on to ask some behavioral and situational questions. Asked current comp and expected comp?? Idk if managers can ask, but whatever.

I reached out to recruiter next day and she said Hiring manager rejected and feedback was "My knowledge in skills was not as expected and did not seem culturally fit to work in the environment"

Duration: 90 Mins Coding + 15 Mins Evaluation

PhonePe is conducting its annual hackathon and wants to create an interactive platform to host the event. It will be a 2 day event and contestants need to maximize their score over this period by either solving more problems or more difficult problems. Design the backend to support this platform.

Mandatory Functionalities

The system should have the capability to add problems to the Questions Library. Contestants should be able to register themselves with their name and their department name. A problem should have attributes like description, tag, difficulty level (easy, medium, hard), score. Contestants should be able to filter problems based on difficulty level or tags and sort them based on score (design should be extensible to other attributes ) A contestant should be able to solve a problem as well as get the list of problems solved by him/her. A contestant should be able to see the number of users that have solved a given problem and average time taken to solve that problem. Scoring strategy for a problem could simply be to award the score assigned for the problem or could be something different like a combination of score and time. Return the current leader of the contest Users should be able to get curations like Top 10 most liked problems of a certain tag. Extension Problem

On solving a problem, users should get a recommendation of top 5 problems based on relevance The recommendation strategy for problems could simply be similar tags or extensible to include other factors like number of users who have solved a particular problem or a combination of factors ( Design should be extensible). Capabilities

Below are various functions that should be supported with necessary parameters passed. These are just suggestions, the signatures can be altered as long as the functionalities are provided.

Registering a problem - addProblem() Registering a user - addUser() Fetch the list of problems - fetchProblems() - Should take as input filtering and sorting criterias and return all matching problems in the right order. The result should contain problem attributes like name, tag, difficulty level, score etc. Should also display number of users who have solved a problem and the average time taken for that problem Solve a problem - solve() - Exposed to a contestant to mark a problem as solved [ For the extension problem, this function should return next 5 recommended problems ] Fetch solved problems for a user - fetchSolvedProblems() - Fetch the list of solved problems for a user Get leader - getLeader() - Returns the name and department of the leader Get top n liked problems of a certain tag - getTopNProblems() Guidelines

You should store the data in-memory using a language-specific data-structure. Implement clear separation between your data layers and service layers. Simple and basic function are expected as entry point - no marks for providing fancy * restful API or framework implementation Because this is a machine coding round, heavy focus would be given on data modeling, code quality etc, candidate should not focus too much time on algo which compromises with implementation time Expectations:

Your code should cover all the mandatory functionalities explained above. Your code should be executable and clean. Your code should be properly refactored, and exceptions should be gracefully handled. Appropriate errors should be displayed on the console How will you be evaluated?

Code Should be working Code readability and testability Separation Of Concerns Abstraction Object-Oriented concepts. Language proficiency. Scalability

PhonePe processes a vast number of transactions every day, wherein some transactions may fail (enter a FAILED state) or remain in a PENDING state due to various reasons such as bank or NPCI issues, or internal PhonePe errors. To handle such cases efficiently, a resolution system is needed, where customers can log their unsuccessful transactions and raise complaints against them. The system must categorize customer issues into several types, such as payment-related, mutual fund-related, gold-related, or insurance-related. Different customer service agents will have their specific expertise based on the issue type, whom the system will assign the issues by marking them waiting in case all agents are busy.

The customer service agents can work on one issue at a time and update its status, and once it is resolved, the agent will receive another issue.

The solution should incorporate the following features:

Customers can log a complaint against any unsuccessful transaction. Customer Service Agents can search for customer issues with issue ID or customer details (email). Agents can view their assigned issues and mark them resolved once they are resolved. System should assign the issue to agents based on an assigning strategy. System should allow the admin to onboard new agents. System should allow the admin to view the agent's work history. Implementation requirements Your solution should implement the following functions. Feel free to use the representation for objects you deem fit for the problem and the provided use cases. The functions are ordered in the decreasing order of importance (highest to lowest). We understand that you may not be able to complete the implementation for all the functions listed here. So try to implement them in the order in which they are declared down below.

createIssue(transactionId, issueType, subject, description, email) addAgent(agentEmail, agentName ,List) assignIssue(issueId) // -> Issue can be assigned to the agents based on different strategies. For now, assign to any one of the free agents. getIssues(filter) // -> issues against the provided filter updateIssue(issueId, status, resolution) resolveIssue(issueId, resolution) viewAgentsWorkHistory() // -> a list of issue which agents worked on Example:

createIssue("T1", "Payment Related", "Payment Failed", "My payment failed but money is debited", “testUser1@test.com”);

Issue I1 created against transaction "T1"

createIssue("T2", "Mutual Fund Related", "Purchase Failed", "Unable to purchase Mutual Fund", “testUser2@test.com”);

Issue I2 created against transaction "T2"

createIssue("T3", "Payment Related", "Payment Failed", "My payment failed but money is debited", , “testUser2@test.com”);

Issue I3 created against transaction "T3"

addAgent(“agent1@test.com”, “Agent 1”, Arrays.asList("Payment Related", "Gold Related"));

Agent A1 created

addAgent(“agent2@test.com”, “Agent 2”, Arrays.asList("Mutual Fund Related"));

Agent A2 created

assignIssue("I1")

Issue I1 assigned to agent A1

assignIssue("I2")

Issue I2 assigned to agent A2

assignIssue("I3")

Issue I3 added to waitlist of Agent A1

getIssue({"email": "testUser2@test.com"});

I2 {"T2", "Mutual Fund Related", "Purchase Failed", "Unable to purchase Mutual Fund", "testUser2@test.com", "Open"}, I3 {"T3", "Payment Related", "Payment Failed", "My payment failed but money is debited", , "testUser2@test.com", "Open"}

getIssue({"type": "Payment Related"});

I1{"T1", "Payment Related", "Payment Failed", "My payment failed but money is debited", "testUser1@test.com", "Open"}, I3 {"T3", "Payment Related", "Payment Failed", "My payment failed but money is debited", "testUser1@test.com", "Open"}

updateIssue("I3", "In Progress", "Waiting for payment confirmation");

I3 status updated to In Progress

resolveIssue("I3", "PaymentFailed debited amount will get reversed");

I3 issue marked resolved

viewAgentsWorkHistory()

A1 -> {I1, I3}, A2 -> {I2} Things to keep in mind

Programming Language & Environment: You can use the programming language of their choice to implement the solution. Free to use any local IDEs, such as IntelliJ, or the CodeSignal platform for writing code. Scope of Implementation: Use of REST APIs is not required; implementing equivalent functions for the functional requirements is sufficient. Data Persistence: Persistence should be achieved using in-memory data structures (e.g., Lists, Queues, Maps), and no connection to databases is needed. Third-Party Libraries: It is recommended to refrain from using third-party libraries like Guava or Hystrix for functional requirements. Demonstrating Functionality: There is no need to handle command-line inputs, inputs can be hardcoded within a driver class. A main() class or driver or test cases that simulates the functional requirements is sufficient. Ambiguity in Problem Statement: If there is any ambiguity in the problem statement, you can make reasonable assumptions and proceed with the solution. These assumptions should be clearly called out in comments within the code for them to be discussed in evaluation round. Submission Guidelines: You must zip your solution files and email them to the provided email address (Email address must be available in Interview Invite Link). This must be done even if you choose to code on Code signal Platform. Please ensure you submit your solution on time even if it is incomplete. A partial solution submitted on time is better than a complete solution submitted after time. Evaluation Criteria

Completeness of functional requirements Application of OO design principles Code efficiency Code readability and maintainability Testability Handling corner cases Language proficiency Time Duration You will have 1.5 hours to submit your solution to this problem statement.

[execution time limit] 4 seconds (sh)

[memory limit] 2g

Round 1: Machine coding

1.5 hrs

1. Handle multiple communication request types (Email, SMS, Soundbox, etc.).

2. Different request types have different mandatory input fields (e.g., Email: sender, receiver, subject, message; SMS: mobile number, message)

3. Providers can support one or multiple channels; multiple providers per channel allowed.

4. Each provider can have multiple accounts, segregated for better handling (e.g., critical OTP communication).

5. Randomly select one active provider from eligible providers for each request.

Round 2: PS/DS

1)https://leetcode.com/problems/remove-k-digits/description/

2)https://codeforces.com/problemset/problem/1037/D

Round 3: HLD

Design Instagram

Note: For feed publishing and fetching, Interviewer didn't agree with Message queue approach as it is not scalable for billions of users.

Round 4: HM

Project discussion

Verdict: Selected

Compensation: https://leetcode.com/discuss/post/6592499/phonepe-backend-sde-3-yoe-offer-by-anony-cpxy/

I recently interviewed for a Senior SDE role at PhonePe. The interview process included a Machine Coding Round and a System Design Round.

Machine Coding Round

This round started with the recruiter sharing a Google Meet invite. Once I joined, they announced the question and instructed us to email our solution as a zip file to HR. We were allowed to leave the call once we fully understood the question, and it seemed acceptable to take slightly more than the allotted two hours to submit.

The problem involved designing a system where multiple workflows could be configured. Users perform various actions in the PhonePe app, such as signing up, logging in, checking offers, recharging, or making UPI transfers. A workflow consists of several stages; for instance, an "onboarding workflow" might progress through stages like "KYC pending," "Verified user (with first payment pending)," "First payment complete," and "Eligible for a new user coupon."

Workflows are represented as directed graphs where each node is a stage. As a user performs actions, they transition between stages. For example, completing KYC would update the user’s stage from "KYC pending" to "Verified user (first payment pending)." Some stages could have multiple next stages, depending on the user’s action. The system had to support multiple workflows, and a single stage could appear in several. When the service received an action, it needed to update the stages of all workflows the user was part of and check for new workflows the user could onboard into.

The task emphasized implementing production-grade code, requiring classes for workflows, stages (considering singleton patterns), different action types (following SOLID principles), and a service layer for business logic. After submitting my solution, a follow-up call was scheduled to explain it to the panel.

System Design Round

The system design challenge was to architect a notification service capable of sending email, SMS, and push notifications. A key requirement was handling varying delays for different notification types and ensuring dynamic scalability for specific notification categories.

I had a comprehensive interview experience for the Software Engineer - Android position at PhonePe, spread across five distinct rounds.

Round 1: Online Assessment
This round was an hour-long online test. It comprised 10 Android Multiple Choice Questions (MCQs) and one Data Structures & Algorithms (DSA) question, which was of medium difficulty.

Round 2: Problem Solving/Data Structures
Originally scheduled for an hour, this round extended to 1 hour and 30 minutes. I was presented with two LeetCode medium-level questions, primarily revolving around Graphs and Arrays. I was expected to write fully functional code, test it with various inputs, and finally discuss the time complexities of my solutions.

Round 3: Platform (Android)
This one-hour round began with basic introductions. Following that, the discussion delved deeply into all major Android components, covering their functionalities and interconnections.

Round 4: Machine Coding
For this 1 hour 30-minute machine coding round, an online platform link was shared. The core task was to design a utility function for an Android application to synchronize contacts between the device and the backend. I initiated by asking clarifying questions and outlining the assumptions I would make. I was expected to structure different Android layers and define models necessary for API calls. After completing the basic implementation, the interviewer presented several subtasks, such as managing contacts for multiple devices for a single user, which I then had to address.

Round 5: Hiring Manager
The final round, lasting an hour, started with introductions and a thorough walkthrough of my resume. The interviewer then inquired about challenges I had faced in my previous jobs. Following this, they picked up on some Android technical terms I had used and conducted a deep dive into those topics for about 30 minutes. Subsequently, the interview shifted towards behavioral questions, focusing on my ideal team, issues I might encounter in my current team and company, and strategies for resolving such problems.

A few days after the final round, I was informed of my selection, and the offer discussion took place a week later.

My PhonePe SDE-2 Interview Experience

Hi everyone, I wanted to share my recent interview experience for the SDE-2 position at PhonePe. The overall process was quite comprehensive, although my interaction with the recruiter was less than ideal. She was unprofessional, frequently canceled meetings without informing me promptly, and often scheduled interviews on different days than requested. I often had to call back after each round to schedule the next one, which was frustrating.

Round 1: Machine Coding Round

For the first round, I was given a problem and had 24 hours to complete it. The problem was the classic Snake and Ladder game. I successfully completed this round.

Round 2: DSA/Problem Solving

This round consisted of two problems. The first involved some mathematical calculations, while the second was similar to problems that require equalizing all elements in an array through various operations.

Round 3: System Design

This round was focused on system design, where I was tasked with designing a Subscription Management System. I discussed various aspects of designing such a system, including handling recurring payments, plan changes, cancellations, and notifications.

Round 4: Hiring Manager (HM) Round

The final round was with the hiring manager. This round focused on my past projects, behavioral questions, and situation-based scenarios.

Despite the initial struggles with the recruitment process, I received positive feedback from all rounds and eventually got the 'Selected' verdict!

Round 1: Machine Coding

For the first round, I was given a Machine Coding problem and had 24 hours to complete it. I plan to share the specific details of this problem once all my interview rounds are fully completed.

Round 2: Data Structures & Algorithms (DSA/Problem Solving)

This round consisted of two problem-solving questions. The first problem involved some mathematical calculations. The second problem was similar to the concept of 'equalizing all array elements'.

Round 3: System Design

The third round was a System Design interview. After a few reschedules, an interviewer finally joined, and I was tasked with designing a 'subscription management service'. Initially, the result for this round was pending, but I later received positive feedback for it.

Round 4: Hiring Manager Round

My final round was with a Hiring Manager. The outcome for this round was also pending for some time, but I eventually received positive feedback from the recruiter.

Overall, all my interview rounds received positive feedback, and I am now waiting for the final decision regarding an offer.

I received an email from an HR for an SDE opening at PhonePe, which included the job description and the interview process details. I have 2 years of experience.

Round 1: Machine Coding Round

In this round, I was given a problem and had 90 minutes to submit the code. I could use my personal IDE or the one provided by the interviewer. The question involved designing the backend for an interactive hackathon platform. The focus was on data modeling, code quality, and implementing mandatory functionalities to manage problems, contestants, scores, and leaders, with an extension to show problem-solving statistics.

Round 2: DSA and CS Concepts

This round focused on Data Structures & Algorithms and Computer Science concepts. One question involved using a Trie data structure. Given an array of strings and a search key, I needed to return the top 3 matching strings that started with the search key. I don't recall the other problem, but it was an easy LeetCode type question.

Round 3: Hiring Manager Round

This final round was a discussion primarily centered around my previous projects and experience.

The entire process was completed within two weeks, and I received a positive response. I got a call from HR to discuss compensation, but that call never materialized. This happened because I mentioned I was also awaiting compensation numbers from Adobe, which in retrospect, was a mistake. After 3-4 days, when I contacted them again, I was informed that due to internal movements to fill the requirement, I was being put on hold for the position.

I recently interviewed for an SDE 2 position at PhonePe in Bangalore in August 2023. The interview process consisted of three rounds: Machine Coding, Problem Solving/Data Structures, and System Design. I felt I performed well in all rounds and received good feedback, but was unexpectedly rejected before reaching the Hiring Manager round. I hope sharing my experience can help others.

I had a comprehensive interview experience with PhonePe for the SDE-1 role, which started with an offline campus visit to PEC – Chandigarh in August 2022. The entire process consisted of four challenging rounds.

The first round was an Online Test of 90 minutes on the DoSelect platform, featuring four hard-level coding questions. I faced problems similar to Split Array Largest Sum and a variation of Flower Planting With No Adjacent involving minimum cost. There was also a challenging DP problem about maximizing people passing a test based on scores and bounds, and a unique array permutation problem defining "sadness." I managed to pass 7 out of 10 test cases for the first two and attempted the others. Out of 181 eligible candidates, 18 were shortlisted.

The second round was a Technical Round lasting 60 minutes. After my introduction, the two interviewers focused on DSA. I solved Amount of Time for Binary Tree to Be Infected by explaining both BFS and DFS approaches, writing pseudocode for BFS, and implementing DFS. For Sum of Subarray Minimums, I initially gave a brute-force O(N^2) solution, then optimized it to O(N) using a stack, correcting my code after the interviewer pointed out edge cases. Both interviewers were satisfied. 11 candidates proceeded to the next round.

The third Technical Round also lasted 60 minutes and immediately delved into DSA. I explained the recursive (brute force) approach for Generate Parentheses to the interviewer's satisfaction. For Burst Balloons, I fully explained the DP approach, including the state definition dp[i][j], but couldn't write the complete code, yet the interviewer was content with my explanation. Finally, for Optimize Water Distribution in a Village, I proposed an MST approach for pipes and a power set for wells. When prompted for optimization, I received a hint from the interviewer that enabled me to explain the complete optimal solution. 5 candidates were shortlisted for the final round.

The fourth and final round was a 70-minute HR + Technical Round. It began with company-specific HR questions such as "Why PhonePe?", "Why not Google/Amazon?", and "What made me choose PhonePe?". Following this, we had an in-depth discussion about my E-Commerce Web Application project, which I built using Django REST Framework. I confidently answered all project-related questions. Then, I was given a puzzle: "Find the faster rat among 12 rats using 4 cakes." This was followed by variations, including finding the faster rat among 16 rats and determining the maximum number of rats for which the faster one could be found using 4 cakes. I came up with a solution for up to 36 rats and indicated the answer would be in the 36-48 range. I was also asked CS fundamentals questions like the "Difference between Stack and Heap memory," "Inheritance," "Virtual functions and classes in C++," and "Deadlock conditions and scenarios." I admitted I wasn't prepared for DBMS questions when asked. The round concluded with a system design problem: "Design a system for a single lift with 1000 building floors." After explaining my approach, the problem was extended to "design for 3 lifts" and identify suitable data structures. My initial approach using maps and different classes was not satisfactory, but I revised it using virtual classes and maps, which was accepted.

Ultimately, I was one of 4 candidates selected, marking a significant achievement for me, especially after not securing an internship and getting a placement on the first day of the placement cycle. I'm thrilled to give back to the community!

I recently interviewed for a Senior Software Development Engineer position at PhonePe. My journey began with a call from a consultant, leading me into a multi-round interview process.

My first round was a Machine Coding challenge. I was tasked with creating a cab management system, with a strong emphasis on design principles and modular coding. This round involved an hour-long discussion.

The next day, I received positive feedback that I had cleared the machine coding round and was invited for a System Design interview. In this round, I first had to explain my current project. Following that, I was asked to design a distributed ID generator capable of handling 1 million requests per day. Luckily, I had previously practiced this specific problem during mock interviews, even referencing a resource (link). I walked them through the features list, back-of-the-envelope estimations, and discussed potential concerns and issues. Topics like virtual nodes, partitioning issues, and consistent hashing were also part of the conversation.

Two days later, I got feedback that I was moving to the Problem Solving round. The first problem given was Rotting Oranges. I coded this completely, focusing on modularity, and achieved an optimized time and space complexity. The second problem was Longest Valid Parentheses, which I solved using a stack and in optimized time, explaining my approach with pseudocode. In the last 5-10 minutes, I was given Daily Temperatures. I explained a brute force approach but struggled to arrive at an O(n) solution within the remaining time.

After that, I was scheduled for the Hiring Manager round within 2-3 days. The hiring manager was a very experienced individual. He primarily inquired about my current project and daily work. He then presented a simple design problem: how to design search functionality, asking questions about estimations and suitable technologies.

I was hopeful for a positive outcome, but I was informed by the consultant that they wished for one more PS or System Design round (the exact round wasn't clear as HR conveyed this personally to the consultant). I was ready for it, but the very next day, I received an email stating that they had decided not to move forward with my candidacy. I didn't receive any proper feedback. However, I must say that the interviewing staff was good, and the quality of the interviews was high.

I recently interviewed for a Software Engineer position at Phonepe in Pune, and I'm thrilled to share that I received an offer!

The interview process consisted of three rounds:

1. Online Coding Assessment: This round was conducted on the CodeSignal platform. The questions were similar to those mentioned in this LeetCode post. I was able to successfully solve all the problems.
2. Problem Solving/Data Structure Round (1 - 1.5 hours): I interviewed with a senior software developer at Phonepe. The interviewer was very supportive and provided helpful hints whenever I encountered a roadblock. It was crucial to grasp these hints to progress towards the solution. The questions I was asked were related to classic LeetCode problems.
3. Hiring Manager Round (1.5 hours): My interviewer was an engineering manager. The round began with introductions, followed by a detailed discussion about my current company's projects. We then delved into designing e-commerce websites, similar to Flipkart or Amazon. Finally, I was asked several Java-based questions focusing on OOPs concepts.

Overall, the interview process was extremely smooth. I received an update about my interview schedule approximately three weeks after completing the online assessment. All subsequent rounds were conducted on the same day, and I received the results the very next day.

I had an off-campus problem-solving interview at Phonepe. The overall experience was quite pleasant, and the interviewer was very supportive. We began with standard introductions before diving into the technical questions. I was presented with three distinct problems. The first involved optimizing a restaurant's dish selection based on quality and price, which I couldn't fully solve; I later realized it was a variation of the Longest Increasing Subsequence problem, requiring a specific sorting approach. The second question was conceptually very similar to the LeetCode problem: Minimum Speed to Arrive on Time. For the third question, a dynamic programming challenge about a rabbit maximizing points with limited jumps, I struggled particularly with efficiently tracking the maximum value within a moving window of size k. The interviewer eventually hinted at using a deque for this. Given that I only managed to solve about one and a half questions to their satisfaction, I anticipated not receiving a call back.

My interview process at PhonePe was very smooth, with all rounds concluding within a week. There were a total of three rounds:

• Online Coding Assessment: This was the first step.
• DSA-based technical round: In this round, I was asked three programming questions, which were considered to be of medium-hard difficulty.
• Hiring Manager round: This round focused primarily on discussions about my current company's projects and my understanding of DBMS concepts. A good understanding of my previous company's projects was crucial here.

Although I received an offer, I decided to reject it as I had other competing offers.