Shariar Kabir

Shariar Kabir

AI Safety, Interpretability, Behavioral evaluation of LLMs

Hi There! πŸ‘‹

My name is Shariar (/ʃɑːriΛˆΙ‘Λr/ πŸ”Š), or simply SK if you prefer. I am currently working on using interpretability methods to understand the unreliable and inconsistent behavior of neural networks (particularly LLMs). I am particularly interested in how LLMs’ behave in subjective contexts, how their internal mechanisms can be made interpretable, and how we can validate them causally through targeted interventions.

I am actively seeking PhD and Fellowship opportunities in AI safety, interpretability, and behavioral evaluation of LLMs. If our research interests align, or you’d like to collaborate, please feel free to reach out !

In Spring 2026, I joined SPAR SPAR to work on real-time automated mechanistic interpretability methods for AI safety and interpretability, under the mentorship of Sriram Balasubramanian.

Previously, I was a research intern at the NLP Lab in UC Riverside, under Prof. Yue Dong, where I was also fortunate to work with Prof. Kevin Esterling. I worked on behavioral evaluation of LLMs, and explored how psychometric and Bayesian modeling techniques can quantify and explain complex social behaviors in LLMs.

Prior to that, I led the AI Research and Engineering team at Celloscope Ltd. I worked on inclusive AI systems for low-resource languages, including Bengali medical ASR and document understanding tools.

I completed my BSc and MSc in Computer Science and Engineering from Bangladesh University of Engineering and Technology (BUET) BUET. During my academic career I worked extensively on cloud computing and distributed systems, under the supervision of Prof. Muhammad Abdullah Adnan.

πŸ“° News