Summary
Rafiqul Rabin is an AI research engineer specializing in the safety evaluation of code-focused large language models and neural code intelligence, currently advancing digital safety at UL Research Institutes' Digital Safety Research Institute. With a PhD from the University of Houston, he combines deep academic expertise in probing memorization, robustness, and backdoor vulnerabilities of code LLMs with practical experience training and analyzing models like CodeBERT, CodeT5, and CodeGen. His work spans adversarial program augmentations, interpretable prediction techniques, and unlearning approaches to mitigate trojans—bridging ML research and applied cybersecurity for safer developer tooling. Prior software engineering and IoT experience grounds his research in real-world systems, and his high academic performance (PhD GPA 3.94) underscores a rigorous, experimental approach to AI safety.
10 years of coding experience
3 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, 3.94 (/4.00), Doctor of Philosophy - PhD, Computer Science, 3.94 (/4.00) at University of Houston
Bachelor of Science - BSc, Computer Science and Engineering (CSE), CGPA 3.84 (/4.00), Bachelor of Science - BSc, Computer Science and Engineering (CSE), CGPA 3.84 (/4.00) at Khulna University of Engineering and Technology
Higher Secondary Certificate (H.S.C.), Science, GPA 5.00 (/5.00), Higher Secondary Certificate (H.S.C.), Science, GPA 5.00 (/5.00) at Notre Dame College
Secondary School Certificate (S.S.C.), Science, GPA 5.00 (/5.00), Secondary School Certificate (S.S.C.), Science, GPA 5.00 (/5.00) at Sohagpur S.K. Pilot High School, Belkuchi
Bengali, English