Hi! I'm Justin Vincent. I do contract programming in NYC. My specialties are "making things fast" and building distributed systems. I have a math and algorithms background, and am happy to work in most languages. If you have some challenging problems you would like help with, shoot me an email at
email@example.com and let's chat. Nothing makes me happier than to travel the world to work on cutting edge problems.
Algorithm Shop (Q1 2013 - Present). New York.
I have worked as an independent programmer since leaving the Obama campaign. Some projects that I have been involved with:
- Building databases and a voter API to support the 2014 Libyan Elections. Went to Tripoli for 3 weeks.
- Building libraries for fast protein substructure searching for Schrodinger.
- Building text classifiers and search infrastructure for cir.cl.
Obama For America (Q2 2011 - Q4 2012). Chicago.
I was one of the early tech hires on Obama's 2012 re-election campaign in Chicago.
- Built many of the core components of project Narwhal, including real time numbers aggregation, user activity feeds, reporting hierarchy, and search.
- Optimized algorithms and SQL queries to efficiently scale our project up to election day traffic.
- Worked on post campaign cleanup and documentation.
- Built a large scale person matching system to identify voters across our various databases, either in batch or with realtime queries.
- Built person matching models using machine learning, with training data gathered by an internal tool I created.
Google (Q4 2007 - Q2 2011). Mountain View.
Google was my first job out of college. I loved it there, but am enjoying being an independent contractor even more.
AdWords Payment Fraud
- Built and maintained signals for a large scale machine learning system.
- Owned the problem of realtime duplicate account detection.
- Built a testing framework for a large scale, HIGH volume, distributed system.
- Optimized hot spots in the backend processing pipeline.
Google Books (Internship)
- Worked on signals to detect important quotations in the Google Books corpus.
- Co-inventor on patent US20090055389.
Education (2004-2008). Seattle.
I attended University of Washington where I majored in math and computer science. I studied abroad in Russia (math), Rome (creative writing), and Spain (Spanish lit and language). I also took as many grad classes as I could, and definitely thought I would do grad school until I interned at Google and got sucked in by the poshness. I breifly toyed with doing a music degree (classical guitar), but then failed dictation and sight singing (hard).
B.S. Computer Science
- Coauthor of a FOCS paper with James R. Lee, though professor Lee did most of the big thinking.
- Researched particle filter approaches to WiFi based localization with Dieter Fox. We could find where you were in the CS building to within a few meters just based on WiFi signal strengths.
- Took graduate courses in algorithms, complex analysis, geometric embeddings, brain computer interfaces, robotics, and complexity theory.
- Was a TA and led weekly sessions for discrete mathematics, introductory programming, and digital design courses.
- Did research on numerical modeling of stochastic processes with Stephen Rhode.
- Took graduate courses in complex analysis.
- Participated in a mathematical modeling competition (MCM).
- My most familiar language is Python.
- I'm comfortable using Python's various data and scientific libraries like numpy, cython, pandas, scipy, and scikit-learn (see blog!).
- Have been good at C++ (Google) and Java (Google + OFA) in the past. I have done large projects in both, Python is just more fun.
- Have a broad background in algorithms.
- Definitely know my way around a MySQL database.
- Can hack together quick
d3.jsvisualizations (most my blog posts have a visualization).
I speak conversational Spanish, do math as a hobby, and ran a large model airplane trade show in 2012.
Currently Leveling Up!
Currently (1/29/2014), I am learning about probabilistic programming with Allen Downey's excellent Think Bayes book.