Jariullah Safi
VP Of AI And Computer Vision at Simbe
Round Rock, Texas, United States
Summary
👤
Senior🎓
Top SchoolJariullah Safi is a VP of AI and Computer Vision with 11 years of experience building production-grade perception systems, currently leading AI at Simbe Robotics in Round Rock, Texas. He progressed from hands-on robotics and CV engineering to director-level leadership, establishing CI/CD, scalable pipelines, and bespoke real-time object detection and OCR solutions deployed at scale. Grounded in rigorous system theory and estimation research from grad work at Penn State, he brings deep expertise in state estimation and battery system modeling alongside practical ML engineering. Notable open-source work includes optimizing a CVPR 2020 3D photo inpainting pipeline for a 30% speedup, reflecting his focus on performance and profiling. He blends academic curiosity—preferring the “aha” moments of analytic insight—with product-oriented execution, mentoring teams to turn research ideas into reliable systems. Outside core duties he’s a polymathic maker who also creates technical content, including YouTube videos, underscoring a talent for communicating complex ideas.
11 years of coding experience
10 years of employment as a software developer
Master of Science (M.Sc.) Mechanical Engineering, Master of Science (M.Sc.) Mechanical Engineering at Penn State UniversityBachelors of Science Mechanical and Nuclear Engineering, Bachelors of Science Mechanical and Nuclear Engineering at Idaho State University
English, Urdu, pubjabi
Github Skills (5)
computer-vision10
performance-optimization10
python10
mesh-processing10
pytorch8
Programming languages (17)
Github contributions (5)
vt-vl-lab/3d-photo-inpainting
Apr 2020 - Apr 2020
[CVPR 2020] 3D Photography using Context-aware Layered Depth Inpainting
Role in this project:
ML Engineer
ML EngineerContributions:8 commits, 1 PR, 5 comments in 10 days
Contributions summary:Jariullah focused on optimizing the 3D photo inpainting pipeline. Their contributions include adding time tracking to identify performance bottlenecks in the code, which was followed by optimizations to the bilateral filtering process within the mesh processing stage, and the core `main.py` file, resulting in a 30% speed improvement. Further code updates included streamlining and improving mesh processing operations and removing redundant imports.
context-awarestructure-from-motioncvpr-2020inpaintingcomputer-vision
safijari/yag-slam
Aug 2019 - May 2022
A complete 2D and 3D graph SLAM implementation using plagiarized code from Karto
Contributions:73 commits, 18 PRs, 122 pushes in 2 years 9 months
graph-slamgraph3d-graph
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.