Projects
Selected work from production ML systems, open-source tools, and technical explorations. Each project includes context, architecture decisions, and outcomes.
RAG Pipeline for Enterprise Search
End-to-end retrieval-augmented generation system serving 10K+ daily queries with sub-second latency.
RAG LLM Vector Search Python
LLM Observability Platform
Monitoring and tracing infrastructure for LLM applications — catch regressions before users report them.
Observability LLMs MLOps Tracing
MLOps Platform for Model Deployment
Internal platform for deploying, versioning, and monitoring ML models — from training to production in hours, not weeks.
MLOps Infrastructure CI/CD Kubernetes