Hi there. I’m Nick.
Founder with successful exit · BS Astrophysics · MS Computer Science
Currently focused on AI safety. Grounding myself in the fundamentals and building empirical testbeds to study real agent behavior. Still love crafting secure full-stack web apps. Previously founded a game company, ran it for over a decade, and had a successful exit.
I'm a software engineer with a founder's mindset. I focus on the results that matter. I have 20+ years of web development experience, with deep expertise in full-stack LAMP, Python, PHP, and Perl. I also work in TypeScript and have spent recent years building reinforcement learning projects and environments from scratch with PyTorch.
I founded Game Craftsmen and ran it for 12 years. I bootstrapped it with credit cards, ran a Kickstarter, raised angel and institutional funding, and grew it to profitability. I had a successful exit, every investor made money. The new owner is profitable, my entire team stayed in place (minus me), and the games are still running with a thriving community. Win-win-win.
I hired and managed 20+ employees and 100+ freelancers, establishing all processes across the org. I very much enjoyed mentoring many junior engineers and launching their careers. Working with so many junior engineers has translated extremely well to working with modern AI coding workflows. You need to figure out what context they need and make sure they have it. Tell them which patterns to follow, point to which parts of the codebase are relevant. Give them clear goals, explain what is important, why it is important, and have patience. Sometimes they don't get it right the first time. Good tests and tight feedback loops help.
I've also been CTO at a long running startup, and a software engineer at a company that was acquired by Google for $500M. See resume.
I hold an MS in Computer Science and a BS in Astrophysics from Michigan State University.
Grant to work from Foresight's San Francisco node. Building an open source framework for reproducible multi-agent AI safety research. Fork and test methodology, designed to surface surprising emergent behavior for alignment theorists.
Four week intensive alignment fellowship at Château Hostačov, Czech Republic. Study of Agent Foundations: decision theory, instrumental convergence, Goodhart’s Law, etc. Deep thinking with peers, mentors, and visiting researchers.
An incremental browser game teaching AI safety concepts through gameplay. Built solo. Won Grand Prize at the Future of Life Institute contest (300+ entries).
🏆 Grand Prize · FLI
Open-source experimental testbed for reproducible AI safety research. Menu-driven interface to spin up agent worlds with configurable scenarios, LLM backends, and budgets. Fork-and-test methodology: snapshot running experiments and branch from identical starting states.
Foresight Institute Grantee
Multi-agent RL environment: supplier agents compete with real incentives to cheat; an auditor agent (GRPO) detects fraud from indirect signals only. All agents adapt simultaneously.
A playable deep reinforcement learning demo. The Bellman equation is magical!
A comparative study in pathfinding algorithms and reinforcement learning. A* wins.
Ingests legal documents in multiple formats, indexes with FAISS vector store. Answers always include citations to original source.
One of the games I made has players (Superheroes) engaging in over 10,000 PvP battles per day. Hero Stories feeds the most important battles to an LLM in real time. The system publishes stories based on the real players actual actions, their profile, their histories, etc. The players love it.
Looking for a software engineering role in web development, machine learning, or AI safety where I can build great things with a strong team. Open to remote, hybrid, or on-site in the SF Bay Area.