Hi, I'm Chinmay.

I'm looking for my next team!

I'm Chinmay, a Full Stack Software Engineer specializing in AI Native engineering and Agentic AI system, based in Berlin with 3.5+ years of professional experience, including the last two years at the Hasso Plattner Foundation building agentic AI applications, and workflow automation systems. I've shipped a range of AI tools and applications that teams use in production every day, from intelligent assistants and chatbots to document automation pipelines. One of them cut a review process from 5 to 6 hours down to 15 minutes. Before that I spent two years in DevOps at Deloitte with a zero-error deployment record. I work best when I own things end to end, from architecture through deployment and monitoring.

LinkedIn Download CV
Chinmay Bandekar

Projects

Things I've built. Real problems, real code, running in production.

AI CORA

Production compliance assistant for grantee reviews. 5–6 hours of manual work down to 15 minutes per report; used daily by an 8-person team.

PythonDjangoLangChainAzure
Read the story →

CineMind AI

Movie recommendations through vector search and LLM reasoning. Semantic taste-matching instead of ratings-matrix guesswork.

PythonVectorDBOpenAILangChain
View on GitHub →

Cricket Score Predictor

Live score prediction with classical ML. Random Forest model, interactive Dash frontend, deployed on Hugging Face.

PythonRandom ForestDashHugging Face
View on GitHub →

Fantasy XI Debate Engine

Two LLMs argue about cricket until they agree on a team. A multi-agent system where GPT proposes a fantasy XI and Claude critiques it round by round, until consensus or the debate hits its round limit.

PythonFastAPIReactOpenAIAnthropic
View on GitHub →

GenAI Chat App

A ChatGPT-style assistant wired end to end. Replies stream in token by token, images can be attached and asked about, and markdown renders live — on a stateless API with chats kept client-side.

PythonFastAPIReactOpenAIVercel
Live demo → View on GitHub →