Eric Crawford
Senior Autonomy Engineer - Deep Learning at Skydio
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolEric Crawford is a senior autonomy engineer and deep learning specialist with 16 years of experience building real-time vision systems for drones and large-scale neural simulation tools. At Skydio he designs and deploys efficient, quantized instance segmentation models and novel distillation pipelines that improved re-identification and cut on-device latency by a third, while also creating a scalable semantic search system over tens of millions of aerial images. His research background from McGill and Waterloo centers on unsupervised object discovery and biologically inspired representations, and he has productionized those ideas into 2D, video, and 3D object discovery pipelines. A longtime contributor to the Nengo brain-simulation project, he brings rare expertise spanning CUDA/MPI performance engineering, model distillation, and practical on-vehicle inference for constrained hardware.
16 years of coding experience
5 years of employment as a software developer
Master of Mathematics (MMATH), Computer Science, Master of Mathematics (MMATH), Computer Science at University of WaterlooDoctor of Philosophy (PhD), Computer Science, Doctor of Philosophy (PhD), Computer Science at McGill University
Github Skills (10)
neural-network10
machine-learning10
back-end-development10
ng10
python10
algorithms8
data-structures8
algorithm8
data-structure8
testing7
Programming languages (3)
Github contributions (5)
nengo/nengo
Jan 2014 - Dec 2015
A Python library for creating and simulating large-scale brain models
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:14 commits, 3 PRs, 2 pushes in 1 year 11 months
Contributions summary:Eric primarily contributed to the backend of the `nengo` library, focusing on implementing and modifying learning rules, including PES, BCM, and Oja. They refactored existing code, updated syntax, and fixed plotting functionalities to improve user experience. The user also implemented new features such as the SimVoja operator and addressed issues related to predefined vocabularies in the SPA input module, demonstrating their involvement in the project's core functionalities and machine learning aspects.
pythonnengoneuroscienceneural-networks
e2crawfo/clify
Feb 2017 - May 2020
Contributions:1 release, 18 pushes, 2 branches in 3 years 3 months
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.