Real success is finding your lifework in the work that you love.David McCullough

The best way to predict the future is to invent it.Alan Kay

Hi, I’m Huascar (pronounced wäs-kär), an AI researcher at Charles Schwab’s AI.x group. Before that, I was a Senior Computer Scientist at SRI’s Computer Science Laboratory, and earlier still, a Graduate Researcher at the Software Introspection Lab at UC Santa Cruz, where I worked with Jim Whitehead.

My research focuses on making AI systems behave correctly, consistently, and safely in high-stakes environments. I approach this through multi-LLM collaboration techniques, including LLM Chemistry and multi-LLM consensus estimation for model recommendation and output aggregation, which I have applied to data analysis, specification learning, code generation, and program repair. I combine these with static/dynamic program analysis, AI assurance, AI red teaming, and AI evaluation and benchmarking. I also work on social-cyber security to protect open-source software supply chains from social and technical attacks.

I was the PI on DARPA Social Cyber and have contributed to several DARPA-funded efforts, including DARPA ARCOS and DARPA MUSE. I hold a PhD in Computer Science from UC Santa Cruz (2015) and an MS in Software Engineering from San Jose State University (2006).

My Erdös number is 3: P. Schweitzer, B.D. McKay, and P. Erdös.

Cheers,
Huascar A. Sanchez
huascar dot sanchez at gmail dot com