About me

Short version#

I'm a first-year computer science student at the University of Michigan. Most of what I know, I learned by building something, breaking it, and then working out why. This site is where that record lives: projects that got finished, experiments that did not, and notes on what I was reading at the time.

Right now the thing holding most of my attention is training an AI to race in F-Zero — an experiment in genetic algorithms that has taught me more about writing a good fitness function than about racing.

Education#

University of Michigan — Computer Science, first year

I'm just getting started on the coursework, so most of what's on this site came from building things on my own rather than from a class. I expect that balance to shift, and I plan to keep documenting both.

What I'm interested in#

• Machine learning. Especially the parts where you can watch the learning happen rather than reading it off a chart.
• Robotics and control. The gap between what a model assumes and what the world does.
• Systems and software engineering. Writing code that someone else — usually me, six months later — can still work with.
• Game AI. A surprisingly good testbed, because failure is obvious.

Current focus#

• Getting genuinely comfortable with genetic algorithms and reinforcement learning rather than just familiar with the vocabulary
• Building the habit of documenting experiments while they run, not afterwards
• Reading more primary sources and fewer summaries of them

Longer term#

I want to work on systems that have to operate in the real world — robotics, autonomy, or applied machine learning — where the interesting problems come from the environment being messier than the model. Before that, I want a lot more practice at the unglamorous part: making experiments reproducible and being honest about what the results actually show.

Outside of that#

Still figuring out how much of my time this should take up. Suggestions welcome.