juan castro-varón

var links = [ blog , instagram , threads , github ];



Google - The Nature Conservancy Brazil – Software Engineer April 2023 - Current
London - UK Current
  • Working with The Nature Conservancy Brazil to create a tool to identify illegally harvested wood from the Amazon rainforest, using AI to combat deforestation.
Google – Software Engineer March 2019 - Current
London - UK Current
  • Working on the Google Play Developer Console, enabling millions of Android developers worldwide to publish apps to >3 billion devices and grow their business with Google Play.
  • Designed and developed a pipeline to parse deep links in apps uploaded to Play Console, to flag misconfigured deep links. This resulted in 157 apps fixing 1.5M misconfigured URLs so far, delivering $15.4M in ARR in the 2.5 months since launch, projected to generate $70M in 2023, through 2x improved ad and app-purchase conversion rates in-app rather than in-browser.
  • Led the rewrite of the UI components library (used by >200 London-based engineers every week) to support responsive design principles, leading to an increase in mobile user share of the Play Console from 12% to 22%.
  • Rewrote Play Console’s email engine, successfully automating away the effort required to send email notifications (reduced a 8-step process taking several days of engineering effort to just 1 line of configuration per message category).
  • Owner of Play Console's messaging infrastructure, used by >200 engineers every week.
  • Served as manager to two Engineering interns working on the Google Play Console, who both performed above expectations and have received and accepted offers to return to Google in London.
  • Volunteering to provide and enhance opportunities at Google for engineers from under-represented groups in technology. This has included serving as a 1:1 coach and mentor to several engineering interns, creating “Spotlight on London” videos encouraging students in EMEA to choose London as the best place for a technology career at Google, and volunteering at trainings and events.
Google – Software Engineering Intern Feb 2017 - Jun 2017 & May to Aug 2018
London, UK (2018) – Seattle, USA (2017)


Universidad de los Andes | Bogotá, Colombia 2014 – Dec 2018
Ranked #5 top university in Latin America, #1 in Colombia by QS World University Rankings
Bachelor of Science, Systems and Computing Engineering GPA: 4.24 / 5
Mozilla Open Leaders | Remote 2019
Invited by Mozilla to participate as web developer of Citapress.org
Part of the 7th Round of Mozilla Open Leaders, a program by Mozilla to fuel the Internet Health movement, designing and building open source projects that empower collaboration within inclusive communities.


UK Exceptional Talent Visa Holder
UK Home Office – Endorsed in the field 'Digital Technology'
Apple WWDC Scholarship winner
WWDC 2018 Conference in California - Apple Inc.
Twice Linux Foundation Scholarship winner
Open Networking Summit & Open Source Summit Europe
Highest national score in standardized college admission test ICFES
Colombian Ministry of Education
Full-tuition Scholarship winner
Bachilleres por Colombia Program, Ecopetrol S.A.
Highest undergraduate admission score
Alberto Magno Award, Universidad de los Andes


Cita - Online press website (2018 - Current) citapress.org
NYC Harbor Water Quality – Web app (2018) nycwater.castrovaron.com
  • Developed and deployed a web interface for exploring NYC water quality metrics using IDW interpolations.
  • Built using the Google Maps API, showing over >10K samples of water quality data over a span of 100 years on an interactive online map of the NYC Harbor.
Periplo Journal – iOS App (2017) periplo.castrovaron.com
  • Designed, developed and published a native iOS journaling app, with a simple, easy-to-use interface and Markdown support for writing and previewing. Built using Swift and native iOS frameworks.
Become a curator – Random art statement generator (2016) bac.castrovaron.com
  • Developed and deployed a web app (front-end and back-end) for a random art statement generator, using Markov chains to generate clear but meaningless sentences based on large bodies of text.
  • Built using the Flask / Python for the backend, and Angular.js for the front-end. Deployed on Heroku.
  • Achieved > 10K page views and statements generated in the first month.