Hi, I’m Daniel Baranov
I’m an MS student at WPI currently doing my master’s thesis in Mechanistic Interpretability.
Previously, I was a SWE Contractor and Intern at AMD
In my free time, I like to explore research ideas, follow technical rabbit holes, and work on projects driven by curiosity rather than outcomes.
My research interests:
Mechanistic Interpretability, AI Safety, Agentic AI, Backdoor
Attacks and Adversarial Robustness in LLM Agents