Christie Purackal
Computer Vision Engineer at Audioscenic
Southampton, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolChristie Purackal is a computer vision and machine-learning engineer with 11 years' experience focused on real-time pose estimation and SLAM, blending rigorous geometry, optimization, and state-estimation theory with production deployment across heterogeneous hardware. As a Kornia maintainer and mentor in the Google Summer of Code program, she contributes to a widely used geometric-vision library—authoring quaternion and Lie-group functionality that improved performance and robustness. At Audioscenic she rebuilt legacy pipelines into lightweight, real-time CNN solutions for consumer devices and created an internal ML experiment-management platform and a C++ inference runtime targeting CPUs, NPUs and ARM SoCs. Her prior work includes dense visual-inertial SLAM in ROS2 with CUDA/OpenGL and practical calibration tooling, reflecting a strong end-to-end systems mindset from data generation to embedded inference. Based in Southampton, she combines research-grade mathematical depth with hands-on engineering that drives measurable product and performance wins.
11 years of coding experience
6 years of employment as a software developer
Nalanda Public SchoolBachelor's degree, Information Technology, Bachelor's degree, Information Technology at Cochin University of Science and Technology
English, Malayalam
Github Skills (11)
computer-vision10
pytorch10
deeplearning-ai10
deep-learning10
python10
quaternions10
pytest9
algorithms8
data-structures8
algorithm8
data-structure8
Programming languages (6)
Github contributions (5)
kornia/kornia
Sep 2022 - Jan 2023
🐍 Geometric Computer Vision Library for Spatial AI
Role in this project:
ML Engineer
ML EngineerContributions:82 reviews, 64 commits, 35 PRs in 4 months
Contributions summary:Christie primarily focused on implementing and refining functionality related to quaternions within the Kornia library, a geometric computer vision library. They addressed bug fixes, notably a quaternion index issue, and introduced new features such as methods for axis-angle representation, spherical linear interpolation (slerp) and right/left Jacobian implementations. The commits involved modifying and adding tests to support the quaternion-related operations and liegroup functionalities, including improving performance.
pytorchdifferentiablepythonvisiondeep-learning
cjpurackal/m2det-tf
Mar 2019 - Sep 2022
Tensorflow 2.0 implementation of m2det
Contributions:48 commits, 6 PRs, 42 pushes in 3 years 6 months
tensorflow-2-0tensorflow-2tensorflowtensorflow2
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.