Backend Engineer — building distributed systems, AI infrastructure, and developer tooling. Currently Technology Summer Analyst @ Citi.

• Building — Enterprise Run Comparison Platform @ Citi
• Learning — Kubernetes · Go concurrency
• Reading — Designing Data-Intensive Applications
• Status — Available for backend engineering roles

Reliability over cleverness. Observability over assumptions. Automation over repetition. Simple systems scale better. Good APIs disappear.

Jun 2026 – Present

Problem — Manual, error-prone reconciliation between consecutive enterprise feed runs. Solution — A schema-aware Parquet comparison engine with zero-copy streaming, parallel loading, and hybrid caching. Impact — 200K+ records · 8-worker parallel loading · 6 modules · 25+ automated tests

Feb 2026 – Jun 2026

Problem — Slow, N+1-bound document processing with no real-time feedback. Solution — Async FastAPI, Redis and Celery pipeline with batch aggregation and Gemini extraction via Vertex AI. Impact — SSE task streaming · HTTP 429 retry/backoff · RBAC/JWT · 28 rule functions

FastAPI · Postgres · Prisma · Docker · LLMs

Automated email ingestion, interest-based filtering, and calendar sync — Google OAuth, idempotent job queues, and a containerized async LLM pipeline. View repository →

Go · Postgres · Docker

Served 1,200 concurrent users at 99.9% uptime; improved API response times by 35% via query optimization, indexing, and connection pooling; JWT refresh tokens with OTP email verification. View repository →

GitHub · LinkedIn · LeetCode · Email