Summary
Aaquib Syed is a versatile CS and Math student researcher focused on AI interpretability and scalable methods to attribute and remove harmful behaviors in large language models, with seven years of practical experience across research and engineering roles. He has interned and contracted at top AI labs and platforms—Google DeepMind (incoming), Anthropic, and Google Ads ML—building evaluations, LLM capability forecasts, and automated optimization pipelines. Aaquib pairs this research depth with production-scale engineering from Databricks’ Files team and a neural recommendation system at Capital One, demonstrating comfort with distributed systems that handle billion-scale operations. He’s also experienced in quantitative trading and mechanistic interpretability through programs at Stanford’s ERI, ARENA, and Jane Street, bringing probabilistic rigor to ML problems. Notably, he treats experimentation like software engineering—tinkering in separate Git branches while publishing papers—reflecting a practical, reproducible approach to high-risk research.
6 years of coding experience
1 year of employment as a software developer
Regents Diploma with Advanced Designation, Regents Diploma with Advanced Designation at Smithtown High School-West
Bachelor of Science - BS Computer Science Mathematics, Bachelor of Science - BS Computer Science Mathematics at University of Maryland