Jason Kroll

Experience

• 

  Linux Journal

  Jan 1997 - Jan 2000

  I was responsible for verifying the correctness of all technical content in the magazine as well as for performing product reviews. I happened to catch the boom and bust. I checked technical articles for correctness and style, mostly catching typos and formatting errors with the occasional misconjugation thrown in.

  • Technical Editor

    Jan 1999 - Jan 2000

  • Copy Editor

    Jan 1997 - Jan 1999
• 

  ClayValet, Inc.

  Jan 2007 - Jan 2008

  Software Developer

  At ClayValet we built a mixed human and artificial intelligence personal shopping service in Ruby on Rails using natural language processing and Amazon's Mechanical Turk. The system accepted complex natural language queries and returned a detailed report about your purchasing options. I was responsible for the workflow system and the merchant reliability layer.

• 

  IMedExchange

  Jan 2009 - Jan 2010

  Lead Software Developer

  iMedExchange was a secure social network for physicians. I came in to replace an entire dev team that was let go, and I quickly turned the site around. We were able to bring in a new team and successfully transition from Ruby on Rails and MySQL to a Java/Spring/Tomcat setup using the graph database Neo4j to provide maximally relevant feeds tailored to each physician.The neatest thing I did was crawl the web to assemble a detailed database of almost every physician in the country. This was a tricky problem because it was full of duplicates and very noisy! But once uploaded into a graph database it actually worked to produce relevant news feeds. Show less

• 

  Inside Jobs

  Jan 2010 - Sept 2013

  Senior Software Engineer

  Inside Jobs connects careers with education. We use Ruby on Rails, Heroku, PostgreSQL, MongoDB, AWS, and GitHub. After launching the site in my first year, I have done all the dev work for the last 2 years. Here are some highlights:I built a self-generating job sorting and filtering system. Given a seed word like "social", the system churns away on our databases using statistical analysis and natural language processing to assign a "social" score to all jobs. We did this also with "meaningful" and thereby found out what makes a job meaningful (hint: autonomy and social bonds). Currently about 100 filters are live on the site.We developed a system for connecting over 7,000 post-secondary schools in the US to the 15,000+ careers in our system, so that you can find out for any given career what schools in your area can prepare you.We crawled the web for resumes and job listings to identify good stepping-stone jobs to get into a given career and to figure out where one might move up from a given position. Job listings, more so than resumes, help us to identify for different jobs which skills are most in demand by employers. Show less

• 

  Socrata

  Nov 2013 - Jul 2016

  Software Engineer

  We built an excellent Analytics and Machine Learning team.

• 

  Amazon Web Services

  Oct 2016 - now

  Applied Scientist

  AWS DynamoDB Performance