Production systems that handle real scale. Currently at BCode.io building distributed backend services on AWS. Previously at Hack For LA where I reduced API p95 latency 85% and operated Kubernetes clusters at 99.5% uptime.
Heliox — Multi-tenant GPU orchestration platform → 100+ concurrent training jobs · event-driven · <150ms p95 latency → Architecture: FastAPI + Redis + async workers + horizontal microservices
Nova Voice AI — Real-time voice assistant → FastAPI + Whisper + LLM orchestration · Calendar/Gmail API integration → Production containerized with fault-tolerant concurrent request handling
RAG Chatbot — Production-ready retrieval-augmented generation → ChromaDB vector search · OpenAI embeddings · FastAPI conversational API
→ MS Computer Science @ Illinois Institute of Technology (May 2025) → Open to Backend / Platform / Infrastructure Engineer roles → [email protected] · linkedin.com/in/sarishrchavan



