Hridesh Rajan

Papers

Total Citations

H-Index

About

Hridesh Rajan is a leading researcher in software engineering, with a particular focus on the testing and reliability of emerging AI systems. His work bridges the gap between traditional software testing and modern machine learning, especially in the domain of Deep Reinforcement Learning (DRL). A standout contribution is his development of μPRL, a pioneering mutation testing pipeline designed specifically for DRL agents. This work addresses a critical need: as RL is deployed in high-stakes environments like autonomous driving and robotics, ensuring these agents are thoroughly tested before deployment becomes paramount. By introducing a fault-based testing methodology grounded in real-world faults, μPRL provides a rigorous framework for assessing the adequacy of test suites for RL systems. While his most-cited paper has garnered early attention with 4 citations, reflecting the novelty of the field, Rajan’s broader impact is evident in his sustained contributions to software testing, program analysis, and language design. His research is essential reading for anyone interested in the intersection of software engineering and artificial intelligence, offering practical tools to build safer, more reliable autonomous systems.