Devansh Tayal

I work on ML security from a systems angle: whether a security claim still holds after a model is connected to tools, state, and real workflows. My current work covers agentic-system boundaries and machine-unlearning evaluation, and I still build developer tools and full-stack products when the problem calls for them. I care about making technically honest things that remain useful to the people living with them, and I'm always eager to learn from their feedback. At home, I enjoy spending time with my family, especially my lab, who's always by my side. I also love connecting with people from different perspectives, watching movies and shows, listening to music, and building new projects.