
Generative AI Engineer
Yoho Technologies Pvt Ltd · Jan 2026 – Present
- Project: Secure Enterprise Knowledge Assistant | Agentic RAG & Generative AI
- Project Highlights:
- • Developed an end-to-end Agentic Retrieval-Augmented Generation (RAG) application using Python, FastAPI, LangGraph, and Ollama for context-aware question answering over enterprise documents.
- • Designed and implemented stateful LangGraph workflows with conditional routing, query rewriting, document relevance grading, and self-corrective retrieval loops to improve response reliability.
- • Built a hybrid retrieval pipeline using Qdrant, combining dense semantic search and sparse keyword search with reranking to enhance document retrieval relevance.
- • Implemented answer grounding and citation verification to identify unsupported LLM-generated responses and improve factual consistency.
- • Integrated NVIDIA NeMo Guardrails as a separate microservice for input/output validation and LangSmith for LLM tracing, debugging, and workflow observability.
- • Developed an offline RAG evaluation pipeline using Ragas to assess faithfulness, answer relevancy, context precision, and context recall.
- • Containerized application services using Docker Compose and implemented GitHub Actions CI with pytest and Ruff for automated testing and code-quality validation.
- Tech Stack: Python | FastAPI | LangGraph | LangChain | Qdrant | Ollama | NeMo Guardrails | LangSmith | Ragas | Docker | GitHub Actions | pytest | uv

