Avinash Kumar Singh

Avinash Kumar Singh

Senior Researcher

About

Dr. Avinash Kumar Singh is a Senior Researcher at Domyn and an AI expert with over 14 years of experience spanning research, engineering, and leadership in artificial intelligence. His work focuses on large language models (LLMs), computer vision, and human–AI interaction, with a strong emphasis on building scalable, real-world AI systems. He has led and contributed to impactful projects in conversational AI, financial reasoning with LLMs, video intelligence, and assistive technologies, combining deep technical expertise with hands-on experience deploying solutions across cloud, edge, and robotics platforms.

Dr. Singh holds a Ph.D. in Computer Vision and has completed postdoctoral research in human-robot interaction at Umeå University, Sweden, and Montpellier University, France. He has contributed to several top-tier conferences and journals and is an active educator, mentoring professionals in areas such as computer vision, natural language processing, and agentic AI systems. His research interests include Responsible and Sovereign AI, with a particular focus on improving the reliability, safety, and behavior of large language models, bridging the gap between theoretical advances and real-world deployment.

Text-to-SQL is, by most leaderboards, a solved problem. Models post impressive scores and the progress feels real, and then the same architecture gets pointed at a real enterprise warehouse and the numbers fall off a cliff.
Why? The benchmarks we celebrate are built on small, tidy, well-documented schemas with neatly phrased questions and a single obvious answer. Reality is the opposite: hundreds of tables with cryptic names, business terms that map to no column directly, joins that depend on knowledge living only in someone's head, and questions where the same word means three different things depending on who's asking. The way we score these systems doesn't survive the move either, a result that's subtly wrong gets treated the same as one that's obviously right, so the failures that hurt most are exactly the ones our metrics are blind to.

This session looks at where the gap actually lives: in the questions we test against, and in the scores we trust. We'll unpack why ""it works in the demo keeps turning into it broke in production, and what better evaluation could look like.

Read More →
View All Speakers

We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our Privacy Policy & Cookies Policy.

Show details