At Turing, I develop AI coding-agent evaluation frameworks and multimodal safety testing, including automated checks for generated patches and tests of prompt-injection and adversarial image scenarios. For client work, I contribute to Apple Intelligence by implementing function-calling and MCP integrations, and I’ve worked on improving Llama’s reasoning and code generation workflows for META.
At Tata Consultancy Services, I developed machine-learning pipelines to detect false positives in Polyspace Bug Finder and built a web-based graph analysis platform with React and Django. In personal projects, I’ve built an agentic RAG platform for HR search and an AI pull request review system using LangGraph, FastAPI, and PostgreSQL with pgvector.
