At QuantXpress, I designed Python backend services for real-time market data, instrument metadata, and option-chain information, using REST APIs, WebSockets, and asynchronous programming. I also implemented high-frequency data pipelines with PostgreSQL, TimescaleDB, and Redis, and containerized backend infrastructure with Docker for AWS deployment.
I built a Real-Time AI Assistant with Safety-Critical Logic using FastAPI and LLM APIs, applying prompt engineering and context-aware sentiment analysis; the project reports an 87% improvement in response accuracy. For an AI-driven meeting automation project, I integrated Whisper ASR and Groq LLM APIs to turn recordings into structured output, with asynchronous processing and persistent storage.
