Work

Here lies a quasi-accurate timeline of my notable professional moments

Featured Stuffs

Fairy guide
Paper Trails interface

Paper Trails

Few thousand users

A Goodreads for academic papers, blog posts, Lesswrong, everything. Save papers into collections of your choice, leave notes, reviews, and read more!
AidanBench paper cover

AidanBench

NeurIPS '24

Benchmark for evaluating novel idea generation in LLMs using open ended questions (NeurIPS ’24; co-first author).
CASI website screenshot

CASI

Founder

Founded Carnegie Mellon’s AI Safety & Alignment org; hundreds of people, lots of cool AI Safety and AI Governance research.
Virtual Harvard website screenshot

Virtual Harvard

150k+ users

Project I cofounded and built from scratch, acquired by the Harvard University system. This is a to scale campus replica in a Unity & C++ virtual platform.
Multi-Agent RL Trading dashboard

Multi-Agent RL Trading

GitHub

A truly unserious project; this is multi agent reinforcement learning for trading simulations. Explore the repo for training metrics and analysis dashboards.
LLM Explainability dashboard visualization

LLM Explainability

GitHub

A pretty, interactive visualization for embeddings and attention to help interpret model behavior.
Explore
Anthropic ML Research Engineer, Automated Alignment
Paper Trails Solo project; a Goodreads for academic papers, few thousand weekly active users ↗︎
Optiver Quantitative Researcher, led research on superintelligence's impacts on the future
Graduation! I obtained a Bachelors of Science from Carnegie Mellon. Go Tartans!
Anthropic Claude Campus Strategist & Co-Creator
Senators Met 2 senators & congressional staffers; presented on AGI, benefits, failure modes
AidanBench NeurIPS ’24, Co-First Author ↗︎
Cooperative AI Policymaking Platform arXiv, Co-First Author ↗︎
Towards Transparent, Sustainable AI Sole Author, Position paper ↗︎
Amazon Software Developer, made a shadow testing interface for pricing ML models
Multi Agent RL Trading System PPO multi-agent portfolio optimization ↗︎
Interactive LLM Explainability Dashboard Embeddings + attention visualizations ↗︎
METR & CHAI (Center for Human Compatible AI) AI Safety & Risk Researcher
DE Shaw and Co Software Engineering and Quantitative Trading Fellowship
Continuum Space Systems and NASA Software Engineering for satellite systems
MetaBull AI and Disney Lead Developer for a script to video animation model, acquired by Disney
Carnegie Mellon AI Safety Initiative (CASI) Founder & President ↗︎
Key Software Solutions and Apple Machine Learning Engineer (Apple subsidiary)
SureStart & Affectiva AI/ML & Deep Learning Intern (MIT Media Lab)
Virtual Harvard Co-Founder, built from scratch. Acquired by Harvard (2020–2023)
Apple Engineering and Technology Program
NASA Jet Propulsion Laboratory Aerospace Scholar in Math, Engineering, Programming