I build LLM-powered systems that retrieve, reason, and generate grounded outputs from live data.
My work focuses on:
- Agentic AI workflows
- RAG (retrieval-augmented generation) systems
- LLM orchestration + evaluation pipelines
- Backend-driven AI applications
Multi-agent system for automated research workflows
(Planner β Search β Scrape β Retrieve β Writer β Evaluator)
- Processes 5β8 live sources per query
- Uses FAISS + MiniLM for semantic retrieval
- Generates citation-aware, grounded outputs
- Designed with modular, production-style architecture
β https://github.com/TVANISH002/Agentic-Research-Platform
LLM + RAG system for generating personalized cold emails
- Matches job descriptions with portfolio using embeddings
- Uses ChromaDB for semantic retrieval
- Built with LangChain + Groq
β https://github.com/TVANISH002/job2mail-ai-cold-outreach
Document-based QA system
- Indexed 10K+ chunks
- Improved retrieval relevance (65% β 82%)
- Reduced hallucinations through grounding
β https://github.com/TVANISH002/Enterprise-RAG-AI-Assistant
LLM Systems β Agentic AI, RAG, Prompting, Evaluation
Retrieval β FAISS, Embeddings, Vector Search
Backend β FastAPI, APIs, System Design
ML/Data β Pandas, NumPy, Modeling workflows
Infra β Docker (basic), CI/CD concepts
Currently focused on building:
- more robust agentic systems
- stronger evaluation pipelines
- practical production-ready AI workflows
Portfolio β https://tvanish002.github.io
LinkedIn β https://www.linkedin.com/in/anish-tv
Medium β https://medium.com/@anish9