I design, build, and ship production-grade AI systems - not just models, but full-stack products that scale, perform, and deliver real-world impact. My work spans the entire AI lifecycle, from training foundation models and building multimodal pipelines, to orchestrating distributed systems, deploying retrieval-augmented agents, and designing seamless user experiences. I focus on turning AI innovation into reliable, scalable, and usable products.

Core strengths:

• Python & FastAPI for robust AI pipelines, APIs, and microservices

• PyTorch & JAX for LLMs, diffusion, and multimodal models in research and production

• Distributed GPU orchestration using AWS, Modal, and custom worker clusters

• Retrieval & agent systems with vector databases, embeddings, RAG workflows, and hybrid search

• TypeScript & React for high-performance, AI-powered frontends and interactive applications

• Autonomous agents & orchestration frameworks (LangChain, LlamaIndex, MCP-enabled workflows)