Patrick Snape
Research Scientist (Principal) at Meta
Zurich, Zurich, Switzerland
Summary
🤩
Rockstar🎓
Top SchoolPatrick Snape is a principal research scientist and seasoned computer vision engineer with 13 years of experience building production face, eye, hand and body tracking systems for mobile and XR platforms. Currently leading encoding and tracking across Meta’s XR Codec Avatar and Interactions org, he has driven end-to-end delivery from data pipelines that processed hundreds of billions of images to on-device blendshape and photorealistic codec avatar solutions. His background spans Apple (Eye Contact, Animoji/Memoji) and academic work at Imperial College (PhD), combining deep research with hands-on productionization and hardware sensor strategy. An active open-source contributor, he’s helped harden foundational projects such as dlib, OpenCV contrib and conda-build—often tackling cross-compiler and build-environment reliability issues that quietly enable large teams to ship. Colleagues describe him as a pragmatic technical lead who builds teams and tooling that turn cutting-edge research into robust, scalable products.
13 years of coding experience
12 years of employment as a software developer
Doctor of Philosophy (PhD) Computer Vision and Pattern Recognition (Computer Science), Doctor of Philosophy (PhD) Computer Vision and Pattern Recognition (Computer Science) at Imperial College London
English
Stackoverflow
6,336reputation
1.5mreached
176answers
3questions
facebook-graph-api
top-5%
facebook
top-1%
ajax
top-5%
java
top-5%
github-api
top-5%
jar
top-5%
Github Skills (48)
debug10
conda-build10
javascript10
opencv10
c-language10
python10
css10
machine-learning10
user-interface10
firefox-addon10
chrome-extension10
facebook10
google-chrome-extension10
html10
object-detection10
Programming languages (12)
Github contributions (5)
davisking/dlib
Dec 2014 - Feb 2016
A toolkit for making real world machine learning and data analysis applications in C++
Role in this project:
ML Engineer
ML EngineerContributions:29 commits, 2 PRs, 14 comments in 1 year 2 months
Contributions summary:Patrick contributed to the dlib library by adding and modifying code related to object detection and shape prediction, specifically focusing on Python bindings. The commits demonstrate the development of wrappers for shape predictors, including training, fitting, and testing functionalities. Additionally, the user refactored parts of the object detection code, including removing `rgb_pixel` and adding a `drectangle` class and associated methods, thereby extending and improving the usability of the library's object detection capabilities in Python. The user also made several code adjustments and improvements to the existing examples.
data-analysispythondata-sciencedeep-learningc-plus-plus
opencv/opencv_contrib
Aug 2015 - Jan 2016
Repository for OpenCV's extra modules
Role in this project:Back-end Developer
Back-end DeveloperContributions:10 commits, 2 PRs, 9 comments in 5 months
Contributions summary:Patrick primarily contributed to fixing compatibility issues related to specific compilers (MSVC) and standard library implementations within the OpenCV_contrib repository. Their work involved adapting code to handle the absence of certain features like `vector.data()`, `round()`, `pow()` overrides for integers, and `NAN` constants in older MSVC versions. These changes ensured the project's compatibility across different compiler versions, specifically targeting MSVC 2008 and 2010, while ensuring that other compilers are not affected by their edits.
pythoncmakecppcomputer-vision
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.