I build production AI systems — from real-time anomaly detection to LLM pipelines, RAG architectures, and agentic systems that ship and scale. M.S. Data Analytics Engineering, George Mason University · GPA 3.93
From data pipelines and model training to serving layers, vector databases, and production observability.
Production systems with real metrics — not toy demos.
Browser-based voice AI with faster-whisper STT, streaming LLM inference, sentence-buffered Piper TTS, and RAG over a personal knowledge base. The assistant you're talking to on this site.
LangGraph orchestration with 6 specialized agents — document ingestion, retrieval, risk computation, compliance validation, and LLM synthesis. Reduces analysis turnaround from hours to minutes.
Production anomaly detection for solar panel sensor grids detecting early fire risk. Hybrid IQR / Z-score / ROC models with dynamic thresholding, FastAPI backend, PostgreSQL + Redis, Grafana dashboards.
Flask web app automating FDA product code classification from label images using OpenAI Vision API. Batch processing consolidates multiple uploads into single API calls for throughput.
Powered by GPT-4o-mini with RAG over my resume, projects, and Q&A docs. Voice or text — your choice.
Try: "What projects has Chethan built?" or "What's his experience with RAG systems?"
Open to senior AI/ML engineer roles where I own systems end to end. Remote or US-based.