Shaurya Sharma
Building RAG systems, LLM agents & AI backends in Python.
New to Contra
Shaurya is building their profile!
Built an MCP-native AI agent that lets you query any PostgreSQL database using plain English directly from Claude Desktop. You type what you want, the LLM generates the SQL, and it executes against your connected database. Stack: Python, FastMCP, Groq API, PostgreSQL, NeonDB.
1
10
Built a FastAPI service that ingests public Git repos, parses Python code using AST, generates function-level embeddings, and stores them in Supabase with pgvector. Users can ask natural language questions and get LLM-generated answers grounded in the actual codebase context. Stack: FastAPI, pgvector, sentence-transformers, Ollama.
1
15