Raghav Agarwal

I’m an AI Engineer with 2+ years of experience delivering production-grade LLM-powered applications using Python and modern GenAI frameworks. I focus on architecting reliable RAG pipelines and stateful, agentic workflows that turn raw content into trustworthy, actionable outputs.

In my current role, I led Python-based development and deployment of a production GenAI application serving 50+ enterprise clients using FastAPI, MongoDB, and Hugging Face transformer models. I built end-to-end retrieval augmented generation pipelines with FAISS vector indexing (and related tooling), achieving 85% retrieval accuracy and generating factually grounded LLM responses with full source attribution.

I also engineered stateful, multi-step agentic workflows using LangGraph and LlamaIndex, including model-context browser automation for live web research and document extraction—cutting manual research effort by 40%. To improve end-to-end retrieval performance, I implemented autonomous AI agents in Weaviate to orchestrate preprocessing, vector storage, and query workflows, accelerating information retrieval by 30%.

Beyond GenAI, I’ve shipped production web systems and integrations: I delivered responsive financial services web apps and authored/optimized RESTful APIs using Node.js with PostgreSQL while reducing deployment time by 50% through Azure Pipelines. Earlier, I automated CRM operations at Salesforce using Apex triggers and flows, and built Lightning Web Components—then validated my platform depth with 58 Trailhead certifications and 13 advanced super badges.