Raghavendra

A Mind That Never Sits Still.

It all started with maths, a curiosity for patterns, and a fascination with how things work. That curiosity eventually led me to AI, where I found a space to experiment, build, and keep asking questions. Beyond tech, I’m someone who loves exploring new places, discovering different food, meeting people, experiencing cultures, and trying things I’ve never done before. Currently navigating my AI journey through a PGP in AI & Data Science at Jio Institute and an internship at AiSPRY. Always curious, always exploring, and never quite done learning.

See what I've builtGitHubLinkedIn
work

Things I've built

agentic RAG

ClauseCheck

ClauseCheck reads a health insurance claim and decides whether to pay it, and it shows its work. Every clause it relies on comes with the insurer, section and page. When it can't find the evidence it doesn't guess; it hands the case to a human.

The payable amount comes from a plain Python rules engine, because I wanted the same claim to give the same number every time. On a small set of labelled cases it beat a basic RAG setup on verdicts, citations and recall.

  • LangGraph
  • Chroma
  • BM25
  • Groq
  • Ollama
  • Gradio
View on GitHub ↗
machine learning

Autism prediction

A small web app for autism risk screening. You fill in the AQ-10 questionnaire and a few profile details, a scikit-learn model scores it, and the result is saved to MongoDB. The backend is FastAPI.

  • FastAPI
  • Scikit-learn
  • Joblib
  • MongoDB
View on GitHub ↗
full-stack AI app · in progress

AI Job Recommender

I’m building this for Pursue Academia. You upload a resume and it pulls fresh listings from free job APIs, scores every job against your profile, explains the best matches in plain language and emails the good ones, never the same job twice.

The score is the part I care about. The backend works it out from skills, experience, education, semantic similarity and preferences, and the language model only explains the result. It can’t invent a skill or change a score. It’s a work in progress, and I’m testing it piece by piece.

  • FastAPI
  • Next.js
  • Postgres + pgvector
  • Celery
  • Redis
  • Docker
journey

Where I've been

Data science intern, AiSPRYBased in Hyderabad.
NTU SingaporeA two-week immersion on applying AI and digital transformation to real business challenges.
IEEE paper: JANIA personal assistant that runs speech recognition, face login and a local LLM on ordinary hardware, fully offline.
PGP in AI and Data ScienceJio Institute.
Mentor, Pursue AcademiaPart of the gritLOOP Future Skills Path.
Finished my B.E.AI and Machine Learning at DSATM, Bangalore.
AI and ML engineer, Rooman TechnologiesA financial recommendation tool and a phishing detector, built with ML, LLMs and RAG.
Paper on phishing detectionA literature review in IARJSET on how people have approached the problem.
Two ML internshipsA healthcare chatbot at Compsoft, and recommendation models at Bharat Intern.
tools

What I use

Languages

Python · Java · SQL · JavaScript

Machine learning and LLMs

RAG · LangGraph · Scikit-learn · TensorFlow · computer vision · NLP · vector databases

Data and apps

FastAPI · Gradio · MongoDB · Databricks · Delta Lake · PySpark · Git

Certificates

Oracle Generative AI Professional (2025) · Oracle Data Science Professional (2025) · four Google Cloud skill badges in Gemini, prompt design, Streamlit and multimodal RAG

say hello

Let's talk.

If you're working on something interesting, or you just want to talk it through, write to me. I'm happy to learn from people who know more than I do.

raghavshetty1611@gmail.com
Send an emailLinkedInGitHub