Hi, I'm Dristi :D👋

CS 2030 @ UCLA 
          /\_/\
        =( O.O )=
          
          /\_/\
        =( ^.^ )=

I'm currently interested in mech interp and AI safety but exploring the many applications of AI.

  • SPAR Kairos: researching unintended generalization in language models with Dr. Aishwarya Balwani.
  • Kyra Interface: building small language models and tool-calling agents for a privacy-focused personal assistant.
  • A Few Bad Neurons: co-authored research on isolating and correcting sycophancy using SAEs, linear probes, and targeted interventions. Presented at NeurIPS CogInterp, ReliableML, and NEMI.
  • Hack Club: former US West Regional Manager and hackathon director, helping organize events for hundreds of students.
  • Outside of CS, I play piano and flute, practice kung fu, and like building hardware.
email github linkedin resume ↓