Gatlin Johnson

software developer, pug fanatic

Summary

There's not much to it.

I'm a software developer in sunny Austin, TX. Graduated from UT with my BS in Computer Science. I do freelance development and hide in academia. Research interests include machine learning, logic programming, math education, and vodka.

I live in and support cooperatives, play classical guitar, and watch a lot of bad movies.

Hire me!

I enjoy designing and making software. If you want me to make some software, or want to work with me on something, drop me a line. Drop me a line, I say!

Details, details

Projects

Not everything, just the public domain stuff and my shameless plug for Vlad.

Vlad

Vlad is an automatic work / labor scheduling system. It guarantees schedules which assign every job, maximize worker preferences, and which don't conflict with anyone's availability. It's in a pre-alpha stage — sign up for more details!

Saturn

A concurrent network SAT solver written in Go, based on MiniSAT. Pretty simple at the moment. Future plans involve parallelizing SAT instances to take advantage of all those pretty CPU cores you got there.

go-orc

A distributed computing and concurrency calculus, based on work from UT Austin, implemented in Go. Deceptively simple but potentially quite powerful. It was written while I was brainstorming parallelization techniques for Saturn.

CartoCritic

This is a tool created for Texas A&M Corpus Christi for use in the GIS / geospatial engineering classes. It allows students to submit maps for anonymous peer review and grade each other. It's not quite complete - I'm not sure yet how I want authentication to work - but it's handy nonetheless.

Surely

A SAT solving algorithm written in Haskell. This was my attempt at understanding DPLL. I'd like to extend this with clause learning and backjumping, optimize the hell out of it, and put it on Hackage. Hacker News liked it?

HNN

Under major construction!

The Haskell Neural Network library. Under construction! The goal is to produce the fastest neural net library this side of the Mississippi. In particular we will place focus on parallelization.