Summary
Abhiramon Rajasekharan is an AI researcher and Graduate Research Assistant with nine years of experience building interpretable reasoning systems that translate natural language into executable code using LLM agents. He developed REGAL and STAR frameworks that couple finetuned LLMs with logic solvers to produce accurate, auditable decisions in healthcare, business, and conversational AI. His work spans neuro-symbolic methods and code-generation agents, with peer-reviewed publications at venues including WMW, PADL, TPLP, and ICLP. Industry internships at Skyfall AI and AT&T Labs demonstrate his ability to turn research into high-impact systems—e.g., a world model that cut MCTS rollout time by 95% and a fraud-detection algorithm that more than doubled F1. Pursuing a PhD at UT Dallas, he blends deep theoretical grounding with practical engineering to make LLM-driven reasoning both performant and interpretable. Unexpectedly, he often frames reasoning tasks as iterative solver-guided program refinement rather than end-to-end neural prediction, prioritizing transparency alongside accuracy.
9 years of coding experience
1 year of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at The University of Texas at Dallas
Integrated Masters (BTech + MTech), Computer Science, 3.71/4.00, Integrated Masters (BTech + MTech), Computer Science, 3.71/4.00 at International Institute of Information Technology Bangalore