Distributed Systems
Scalable, fault-tolerant architectures that handle millions of requests per day.
Distributed Task Orchestrator
CLIENT · LIVEDesigned and deployed a distributed task orchestration platform handling 2M+ daily workflows across 12 microservices with recursive error recovery.
Database Sharding Layer
CLIENT · LIVEBuilt a transparent sharding middleware for MySQL that horizontally partitioned 8TB across 16 nodes with zero-downtime rebalancing.
Realtime Event Pipeline
CLIENT · LIVEArchitected a Kafka-backed realtime event processing pipeline processing 50K events/sec with exactly-once semantics and sub-100ms latency.
API Gateway Design
OWN PROJECTOpen-source API gateway with dynamic route discovery, rate limiting, circuit breaking, and observability — 1.2K stars on GitHub.
Infrastructure Cost Analyzer
CLIENT · LIVECloud cost intelligence platform that reduced monthly spend by 38% through automated right-sizing, spot instance scheduling, and reservation optimization.
Zero-Trust Auth Mesh
CLIENT · LIVEService-to-service authentication mesh using mTLS and SPIFFE identities across 40+ services with automatic certificate rotation.
Log Aggregation Engine
OWN PROJECTLightweight log aggregation system for edge devices — ingests, indexes, and searches 10GB/day on a single 2GB RAM node.
Network Simulation Toolkit
OWN PROJECTDeclarative network topology simulator for testing partition tolerance, latency injection, and failure modes in distributed systems.
Consensus in Geo-Distributed Systems
RESEARCHPublished paper on optimizing Paxos variants for geo-distributed databases. Reduced cross-region consensus latency by 40% through speculative execution.