Ashish Kumar

I am a machine learning engineer with strong foundations from a top technical institute and hands-on experience across forecasting, ranking, and retrieval-augmented generation systems. I have built and deployed end-to-end ML pipelines using modern tools, emphasizing robust validation, calibration, and scalable inference.

My projects include route-level demand forecasting with validated gradient-boosted models, calibrated offer ranking ensembles for financial offers, and hybrid sequence-to-tree models for fantasy sports prediction; I delivered measurable impact such as revenue uplift and improved ranking metrics.

In internships and hackathons I fine-tuned transformer models, designed RAG pipelines with vector search, and containerized services for AWS deployment—combining rigorous experimentation with production-ready engineering practices.