Summary
Debarghya Datta is an AI researcher and engineer with 8 years of experience, currently a Research Fellow at Microsoft focused on NLP, tool-based LLMs, information retrieval, and dialogue systems. He combines strong academic credentials from IIT Bhilai (MTech) and hands-on industry experience—recently working on SLM/LLM fine-tuning, model distillation, and Indic translation pipelines at Convin. His work includes a publication at EMNLP and active mentorship in the DBpedia Hindi chapter through Google Summer of Code, underscoring a commitment to multilingual and open knowledge projects. Comfortable bridging research and production, he has built end-to-end ML products and QA systems earlier in his career and maintains a GitHub focused on optimizing LLMs. Notably, he brings practical efficiency-focused solutions (distillation, tooling) to scale large models for real-world Indic language use cases.
7 years of coding experience
2 years of employment as a software developer
Bachelor of Technology - BTech, Computer Science, 9.22, Bachelor of Technology - BTech, Computer Science, 9.22 at Kalinga Institute of Industrial Technology, Bhubaneswar
Master of Technology - MTech, Computer Science, 9.72, Master of Technology - MTech, Computer Science, 9.72 at Indian Institute of Technology, Bhilai
English, Bengali, Hindi