- 🎓 B.E. in Computer Engineering, Thapar Institute of Engineering and Technology
- 🏗️ I like designing systems that stay reliable under failure — retries, dead-letter queues, circuit breakers, and observability are recurring themes in my projects
- 🤝 Contributing to open source at DSA-Meet-Design-Pilot — a machine-coding/low-level-design interview-prep resource
- 🌱 Currently exploring AWS Lambda and DynamoDB
- 📫 Reach me at work.gunnishmehta@gmail.com
Distributed Rate Limiter — Node.js Express Redis Lua
Redis-backed rate limiter implementing fixed window, sliding window log, and token bucket algorithms. Deliberately reproduced a real race condition in a naive check-then-act counter, proved it with a concurrency test (50/50 requests bypassed the limit), then fixed it atomically with a Lua/EVAL script (capped correctly at the limit). Also ships as a standalone, npm-installable Express middleware, verified by installing the packed .tgz into a separate project.
URL Shortener + Analytics — Node.js Express PostgreSQL Redis
URL shortener with base62/random short-code strategies and a Redis cache-aside layer (~43% lower latency, ~57% higher throughput under load testing). Async click-analytics pipeline uses Redis Streams consumer groups for crash-safe delivery, with XAUTOCLAIM recovery on worker restart. Consumes the Distributed Rate Limiter above as a real npm dependency rather than reimplementing it.
Distributed Job Scheduler — Node.js Kafka Redis Docker
Distributed job processing system using Kafka consumer groups for horizontal scaling with zero duplicate execution. Implements exponential backoff, dead-letter queues, and idempotent job submission via Redis-backed keys. Benchmarked throughput scaling ~1.87x by adding a worker.
API Gateway — Node.js Express JWT Redis OpenTelemetry
Reverse-proxy gateway centralizing authentication, rate limiting, and routing across backend services. Implements the circuit breaker pattern (closed/open/half-open) for fast failure during outages, with distributed tracing via OpenTelemetry and Jaeger.
Observability Platform — Node.js Kafka Prometheus Grafana
Kafka-based log and metrics ingestion pipeline decoupling event capture from processing. Aggregates service metrics into Grafana dashboards via Prometheus, with threshold-based alerting to reduce mean time to detection.
Sketchy Squad (server) — Node.js Express Socket.io
Real-time multiplayer drawing-and-guessing game built with WebSockets for collaborative, low-latency gameplay.
campus_ride — EJS Node.js
E-cycle booking system for campus use, integrated with Google Sign-In and Razorpay for payments.
Also worked freelance on Estate Flow, a full-stack real estate platform (React Native, Node.js, AWS Aurora, Docker) — client project, not publicly hosted.
Languages: C, C++, JavaScript, Python Backend & Frameworks: Node.js, Express.js, React.js, React Native Cloud & Infra: AWS, AWS Aurora, Docker, Kubernetes, CI/CD (GitHub Actions), Infrastructure as Code Data: SQL, PostgreSQL, NoSQL, MongoDB, Redis Concepts: Distributed Systems, REST APIs, Authentication & Authorization, Scalable Systems, Telemetry, Concurrency & Race Conditions, Lua (Redis scripting) Tools: Git, Postman, JIRA, Linux
Currently interested in Systems/Solutions Engineering roles at the intersection of distributed systems, cloud infrastructure, and automation.

