Open to machine learning engineering roles
I'm David Sasser, a machine learning engineer focused on computer vision and production-ready ML. I care about the full path from data and evaluation to reliable software that people can actually use.
01 / SELECTED WORK
A few examples of how I approach data, model evaluation, engineering tradeoffs, and shipping working systems.
Built a computer vision pipeline to detect and track a baseball through broadcast pitch footage. Fine-tuned a YOLO detector on custom-labeled Baseball Savant frames and used Kalman filtering to maintain a smooth trajectory through missed detections and motion blur.
Achieved 0.94 mAP@50 and 0.66 mAP@50–95 on a challenging small-object detection task. The final system outputs annotated video showing the ball's detected position and tracked flight path frame-by-frame.
A five-round geography game that asks players to recover true north from a rotated, unlabeled satellite map.
Built the scoring, constrained map interactions, accessible compass controls, and animated answer reveal from scratch.
02 / ABOUT
I like turning ambiguous problems into measurable machine learning systems. My approach is practical: understand the real need, inspect the data, establish a sound baseline, evaluate the failure modes, and improve what matters.
I'm especially interested in computer vision work and teams that value rigorous experimentation, clear communication, and strong engineering over unnecessary complexity.
03 / CONTACT
I'm currently looking for my next machine learning engineering opportunity, particularly in computer vision. If my work looks relevant to your team, I'd be glad to hear from you.