Rajeev Vhanhuve

Generative AI Engineer · Machine Learning Engineer

I'm Rajeev Vhanhuve a Generative AI Engineer and ML Engineer based in Pune, India. I work at Infosys, building production GenAI systems for the BFSI space.

Most of my work sits at the messy end of Generative AI and Agentic AI: getting RAG pipelines to behave, wiring up LLM workflows, building citation systems that auditors actually trust, and squeezing latency out of document generation. I shipped Gov Assistant a knowledge tool that lets compliance teams ask questions across thousands of governing documents and get answers backed by real sources and Gov Co-Author, which writes 200-page compliance documents in under 2 minutes using GPT-5.1.

Before GenAI took over my days, I built ML and AI systems at TAO Digital Solutions and Tata Technologies. I've spent a lot of time with Transformers, LLMs, RAG, LangChain, LangGraph, and Azure OpenAI from quick prototypes through to the things that actually run in production.

I care about building AI that is reliable, easy to explain, and useful in places where being wrong has real consequences. Have a look at my projects, read the blog, or drop me a note using the form below.

New post:
GPT-5.1 Unlocked: How I Brought 200-Page Document Generation Down to Under 2 Minutes
March 15, 2026
~ News
[Mar, 2026] ~ $ Gov Co-Author upgraded to GPT-5.1 200-page document generation now completes in under 2 minutes, down from 20 minutes.
[Oct, 2025] ~ $ Gov Assistant officially released on the GenAI portal, serving the Risk & Compliance Tribe with source-backed answers from Policies & Instructions.
[Jul, 2025] ~ $ Custom citation extraction engine for Gov Assistant built from scratch, tracking answers to exact source sections moved to production.
[Mar, 2025] ~ $ Reduced Gov document ingestion latency by 75% after redesigning the chunking and async embedding pipeline.
[Dec, 2024] ~ $ Joined Infosys as Generative AI Engineer to build enterprise GenAI solutions for the BFSI domain.
[May, 2024] ~ $ Promoted to Senior AI/ML Engineer at TAO Digital Solutions; leading end-to-end ML pipeline delivery.
[Mar, 2024] ~ $ Completed Probability & Statistics for ML & Data Science specialization - DeepLearning.AI.
RAG Demo Interactive Query Visualisation

Watch a question travel through a RAG pipeline encoded into a vector, matched against a tiny corpus, and stitched into a grounded answer with citations. Pick a sample query or type your own.

Vector space — cosine similarity
Top-K retrieved chunks
  1. Run a query to see the top matches.
Grounded answer idle

The generated answer with inline citations will appear here.

OFF
ON