RAG systems • MCP servers • Multi-agent workflows (LangGraph) • Full-stack AI products
Open to full-time AI Engineer and AI Full Stack roles • Open to relocate and remote 📧 sonanimansi05@gmail.com • LinkedIn • Portfolio
AI systems that replace manual work, from retrieval to agents to the UI, with live demos.
| Project | What it does | Stack | Demo |
|---|---|---|---|
| Synapse-AI | 5 LangGraph agents that cut B2B API integration from an estimated 3 weeks to 25 seconds | LangGraph, Groq, FastAPI, React 19 | Demo |
| LeadMind MCP | AI lead CRM as an MCP server (8 tools, 2 resources) | Python, MCP SDK, FastAPI, Groq | Demo |
| AgentOps MCP | Autonomous browser agent as an MCP server | Playwright, FastAPI, MCP, n8n, Docker | Demo |
| FlowMind AI | RAG-powered document Q&A | React, FastAPI, ChromaDB, Groq | Demo |
| CampusIQ | Student ERP, 22 modules, 4 portals | React 19, Supabase, Tailwind v4 | Demo |
n8n automations: Immigration Lead Gen • Instagram AI Lead Gen • Lead Management System
AI/LLM: RAG • LangGraph • LangChain • MCP • Groq (Llama 3.3 70B) • Gemini API • Prompt Engineering Backend: Python • FastAPI • Async Python Frontend: React 19 • TypeScript • Tailwind v4 • Vite Data: ChromaDB • Supabase • PostgreSQL • SQLite Automation & DevOps: n8n • Docker • Git • GitHub Actions • Vercel • Render
- BCA, Swami Sahajanand College (First Class)
- Microsoft AI Skills Fest 2025, 50-Day Coding Challenge