Djamahl Etchegaray is a PhD student and machine learning engineer with nine years of experience building and deploying real-world computer vision systems across research and industry. He has a proven track record scaling production ML pipelines—leading a road-defect detection system that processes 260GB of video daily and integrating MLflow, Dagster, and PostgreSQL for automated active learning. His research advances open-vocabulary 3D object detection, including a novel "Find n' Propagate" algorithm that improved novel-object recall by 53% and LiDAR methods that boost AP for unseen classes by nearly 4x. He also brings hands-on MLOps and edge expertise from CSIRO Data61 work that reached 30 FPS on Jetson Orin with major power-efficiency gains and robust online adaptation under distribution shift. Currently based in Brisbane and targeting Summer 2025 MLE internships, he blends academic rigor with production pragmatism and a knack for shipping end-to-end systems from embedded inference to stakeholder-facing mobile apps. An interesting, less obvious strength is his ability to reduce operational noise—evidenced by a custom deduplication algorithm that cut duplicate detections by 90%—helping teams focus on signal, not alerts.
9 years of coding experience
Bachelors of Mathematics / Computer Science, Bachelors of Mathematics / Computer Science at The University of Queensland
Contributions:1 push, 1 branch, 4 comments in 8 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.
Request Free Trial
Djamahl Etchegaray - PHD Student at The University of Queensland