Hey there! I’m Mihir Inamdar — Machine Learning Engineer, data whisperer, and your go-to for turning messy datasets into clean, actionable insights.
I’ve spent the past few years at the intersection of AI research and engineering, working on projects ranging from Generative AI and NLP to computer vision and knowledge graph–powered RAG pipelines. With experience across startups, research labs, and industry, I thrive on building systems that don’t just look good in papers, but actually scale in production.
- Developing end-to-end ML solutions → from data wrangling & feature engineering to training, evaluation, and deployment.
- Productionizing ML pipelines that are robust, reproducible, and cloud-ready.
- Designing GenAI & transformer-based architectures for NLP, recommendation, and multimodal tasks.
- Collaborating across teams (engineering, product, research) to ship AI that makes an impact.
- Bridging research & practice → experimenting with GNNs, LangChain/GraphLang, and RAG optimization.
- An engineering mindset: clean, modular, tested code (goodbye, “it works on my machine” 👋).
- An obsession with metrics that matter — focusing on business value, not just pretty graphs.
- Experience across cloud (AWS, Azure, GCP), ML frameworks (PyTorch, TensorFlow, scikit-learn), and MLOps tools like MLflow, Docker, and CI/CD.
- A love for automating the boring stuff so I can spend more energy on solving the weird, hard, and interesting problems.
Projects, experiments, research implementations, and the occasional proof-of-concept. If you’re into turning data into fuel, models into craft, and production into a beast worth taming — you’re in the right place.
