yash patel

Python, PyTorch, Django, LangChain, RAG, MCP, AI Agents, Multi-Agent Systems, LLM Harness Engineering, PostgreSQL, AWS, and Docker.

Deep diving into LLM inference, CUDA, Triton, GPU architecture, GEMM optimization, memory locality, and efficient model serving.

Building AI-powered applications while experimenting with Transformer internals, PyTorch profiling, GEMM benchmarks, GPU optimization, and LLM inference.

Open to discussions and collaborations on LLM inference, AI systems, RAG, MCP, AI agents, GPU optimization, and open-source projects.