I'm a fresher AI/ML developer who doesn't just build models - I deploy them. Every project I work on ends with a live, usable application. I've shipped computer vision pipelines, RAG systems, and LLM-powered chatbots - all production-ready on Streamlit.

• Currently building AI-powered applications using LLMs, RAG, and Computer Vision
• I use Claude, ChatGPT & GitHub Copilot as development tools - not shortcuts
• Based in Nallasopara, Maharashtra
• Ask me about RAG pipelines, PyTorch, Streamlit deployment, or Groq API
• Reach me at dipankarmane17@gmail.com

Computer vision pipeline for automated fruit freshness inspection

What it does: Classifies 8 fruits as fresh or spoiled in real-time from uploaded images with ~98% test accuracy across 16 classes.

How I built it: 3-phase ResNet50 transfer learning with differential learning rates, automated background removal (rembg), 6-view test-time augmentation, and domain-shift correction by rebuilding the training dataset.

PyTorch ResNet50 Transfer Learning Streamlit rembg Git LFS

RAG-powered research assistant grounded in peer-reviewed literature

What it does: Answers clinical queries using 300 PubMed abstracts as context, citing exact PMIDs for every response - no hallucinations without sources.

How I built it: Full RAG pipeline with custom TF-IDF embeddings, ChromaDB vector store, and Llama 3.3-70b via Groq API. Includes live ingestion stats, article previews, and conversation history.

Llama 3 RAG ChromaDB Groq API PubMed API Streamlit

Multi-chain LLM system with semantic query routing

What it does: Routes user queries intelligently - FAQ questions hit RAG, data questions hit a SQL agent, casual chat hits a conversational chain. No irrelevant responses.

How I built it: Semantic routing architecture dispatching across 3 chains, integrated with Llama 3 via Groq API, with secure environment-based key management.

Llama 3 Semantic Routing RAG SQL Agent Groq API Streamlit

AI & ML

LLM & Gen AI

Languages & Tools

MLOps

• Gen AI & Data Science Bootcamp 3.0 - Codebasics (Apr 2026)
• Gen AI to Agentic AI - Codebasics (Mar 2026)
• Natural Language Processing - Codebasics (Jan 2026)
• Deep Learning: Beginner to Advanced - Codebasics (Dec 2025)
• Master Machine Learning for Data Science & AI - Codebasics (Aug 2025)