I'm starting again at Palantir in July. (LinkedIn won't let me post it with a future date, so I put June. I'll fix that later.)

I joined Palantir to work on AtlasDB (a distributed transactional database). I wrote initial versions of several subsystems, including the garbage collector, migrated legacy software to use the new database, and disentangled code from the legacy codebase in preparation for open sourcing.I founded a team to improve developer tooling and continuous integration on the multi-million line legacy Java codebase at the core of the business. Before I started, builds took hours, on carefully maintained dedicated machines. A dev might find out their change had broken a test weeks after the fact, if ever. The product hadn’t been released in a year. Only a few grizzled experts dared touch the build scripts.When I left the team, every build ran on any developer laptop. Every test was run on every proposed code change within 30 minutes, before the change could even be committed. We effectively moved from “waterfall” to “agile”. The team had, by then, grown to 10 people, two of whom I successfully trained to replace me.I then took over a team, on very short notice, that was working on a performance measurement tool that nobody at the company remembered asking for. I got the project canceled and helped the team refocus on improving the reliability of existing automated performance testing.

After working on TAO (a distributed caching system), I went on to design and implement fbtrace, a distributed RPC tracing system similar to Google's Dapper, which as far as I know was still being used in 2015 and has since been rewritten several times.

System Administration, Experimental Economics, Computer Programming

Teaching assistant for an introductory systems programming class.

For this job, I did grading, administrative, and test automation tasks for a bioinformatics class called "Algorithms for Genomes."

I was a TA for CS24, an intro class on basic computer architecture, operating systems, and distributed systems.

Experimental work in crystal formation.

Experimental work related to the mechanics of suspensions used for LIGO mirrors.

Experimental work related to the mechanics of suspensions used for LIGO mirrors.