Alexander Wan
Undergraduate Researcher
Berkeley, California, United States
Summary
👤
Senior🎓
Top SchoolAlexander Wan is an undergraduate computer science researcher at UC Berkeley specializing in NLP and machine learning, with eight years of hands-on experience in model benchmarking, robustness, and AI policy. He has led and published work on vulnerabilities in instruction-tuned LLMs and LLM detectors (ICML 2023, ACL 2024), demonstrating novel data-poisoning and adversarial attack techniques on large models and training pipelines. Currently at Stanford HAI/CRFM, he focuses on model evaluation and public policy, bridging technical rigor with real-world governance implications. Alexander also builds educational content for ML courses and has scaled experiments on TPUs and multi-GPU systems, reflecting both practical systems expertise and strong research depth. He often explores less obvious angles—such as differences between human and model evidence-resolution behavior in retrieval-augmented systems—highlighting a curiosity for the intersections of ML, security, and societal impact.
8 years of coding experience
3 years of employment as a software developer
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at University of California, Berkeley
Github Skills (9)
language-model7
gaussian-processes6
optimization5
python5
luke4
react3
javascript2
code-generation1
deep-learning1
Programming languages (3)
Github contributions (5)
AlexWan0/infini-gram
Feb 2024 - Dec 2024
An unofficial implementation of the Infini-gram model proposed by Liu et al. (2024)
Contributions:66 pushes, 4 branches, 1 comment in 9 months
AlexWan0/alexwan0.github.io
Jun 2023 - Jul 2026
Contributions:50 pushes, 2 branches in 3 years 1 month
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.