eBay || Staff Software Engineer – AI Platform | Interview Experience
- Posted
- 21 Oct 2026
Summary
I interviewed for a Staff Software Engineer – AI Platform role at eBay and went through multiple rounds covering DSA, system design, AI platform design, and technical leadership. I received an offer with a total annual compensation of ₹85 LPA.
How it went
Base Salary: ₹55 LPA+₹5 LP+₹20 LPA
Total Annual Compensation: ₹85 LPA
1. DSA – Data Structures & Algorithms
Backtracking problems, including N-Queens
- Knight-related follow-up questions
- Graph problems involving BFS, with additional follow-ups
2. System Design – LLD & HLD mixed
Design a Hugging Face‑like SaaS platform
- Apply object‑oriented design principles and design patterns
- Consider multi‑tenancy, API integration, scalability, and testability
3. AI Platform Engineering & HLD
Design a scalable, highly available AI platform
- Model routing, latency optimization, and token‑cost management
- Rate limiting, caching, queues, and asynchronous processing
- Observability, reliability, and cost optimization
4. Staff‑Level Technical Leadership & Engineering Management
Driving architecture decisions across teams
- Handling technical disagreements and architectural trade‑offs
- Mentoring engineers and influencing engineering standards
Questions asked (3)
- 01
N-Queens Problem
Data Structures & Algorithms
Solve the classic N‑Queens problem: place N queens on an N×N chessboard such that no two queens attack each other (i.e., no two share the same row, column, or diagonal). Return all distinct solutions.
- 02
Design a Hugging Face‑like SaaS Platform
System Design
Design a software‑as‑a‑service platform similar to Hugging Face that allows users to host, share, and consume machine‑learning models. Discuss the overall architecture, component responsibilities, data storage, API design, multi‑tenancy considerations, scalability strategies, and how you would ensure testability and maintainability.
- 03
Design a Scalable AI Platform
System Design
Design a highly available AI platform that can serve multiple models with low latency. Cover model routing, latency optimization, token‑cost management, rate limiting, caching, queueing, asynchronous processing, observability, reliability, and cost optimization. Explain how you would handle scaling, fault tolerance, and monitoring.