About me.
Backend engineer.

Software Development Engineer II

Backend services for AI Architect and the AI Code Review Agent.

• Built services that index customer codebases into a cross-repository knowledge graph and serve it to coding agents over MCP. Enterprise users include Swiggy, PubMatic, and Kredivo.
• Reduced LLM operating costs by up to 60% through request batching, changes to frequently used code paths, and lower-cost models, with no measurable loss in output quality.
• Built a CI regression suite that replays pull requests through the code review agent and checks its findings. It replaced manual QA and cut release cycle time by 20%.
• Fixed production bottlenecks in the code review pipeline and compared competing agents to help guide product decisions.
• Won the 2025 company hackathon for the evaluation suite, which was adopted internally.

Software Development Engineer II

Promoted from SDE Intern (Jul 2022) to SDE I (Jul 2023) and SDE II (Apr 2024).

• Built live-class polls, chat, leaderboards, hand-raise, and doubt resolution for 50K concurrent users at 2,000 messages per second. The platform drove 18% conversion from YouTube and a 17% increase in engagement.
• Brought a contractor-run microservice in-house and redesigned its APIs. Request coalescing reduced server and database load by 40% during peak class hours.
• Fixed a JVM bottleneck, removing CPU spikes and increasing microservice throughput by 4×.
• Built search, popular-course, and faceted-filter APIs with Elasticsearch and Redis for the paid content platform.
• Received the Byomkesh Bakshi Award in 2024 for technical contributions and delivery speed.

Research Intern

Research into energy consumption in 5G radio access networks.

• Built ns-3 simulations in C++ to model 5G RAN energy use across deployment scenarios and compare it with theoretical efficiency models.
• Wrote an internal report defining energy-efficiency KPIs for RAN elements and reviewing optimization strategies.