A bit about me

I'm currently a Software Engineer Intern at Locus.sh in Bengaluru, where I own a config-driven automation pipeline for client onboarding on a platform that has powered over a billion deliveries. The pipeline took a 4+ hour manual workflow and turned it into a single API call, and I spent a good chunk of time making it fast: parallel entity migration, dependency-grouped execution, and idempotent upserts pushed throughput up by roughly 90%. I also built the real-time monitoring dashboard that watches it all, streaming live pipeline activity over Server-Sent Events.

Before that, I interned at Bindisa Agritech, building a real-time observability platform with Python, Streamlit, and Prometheus. Watching latency and error metrics across microservices cut debugging cycles by about 25%, and a self-service tool I designed let engineers manage service mesh policies from a UI instead of filing tickets to the platform team. That internship taught me how much leverage there is in tooling that removes humans from repetitive loops, a theme that keeps showing up in my work.

My technical center of gravity is backend engineering and distributed systems, with a growing pull toward GenAI infrastructure. The projects I'm proudest of sit at that intersection: a self-healing system that recovers failed microservices in seconds, and Cortex, a knowledge extraction platform that turns scattered company data into workflows AI agents can safely execute. I care a lot about the unglamorous parts: idempotency, test coverage, failure modes, and what happens when things break at 3am.

Outside of building things, I compete. I'm a Knight on LeetCode, an Expert on Codeforces, and a 5-star coder on CodeChef, with 1000+ problems solved across platforms. Competitive programming is where I first learned to reason carefully about complexity and edge cases, and it still sharpens how I write production code. I'm in my final year of B.Tech at IIIT Allahabad (CGPA 8.57), and I was selected nationwide for Amazon ML Summer School 2024.