Manpreet Minhas
Computer Vision Deep Learning Research Engineer at Towards Data Science
Los Angeles, California, United States
Summary
👤
Senior🎓
Top SchoolManpreet Minhas is a Senior Deep Learning Research Engineer with eight years of experience building computer vision and NLP systems that move from research to real-world impact. He has led projects spanning anomaly detection, segmentation, object detection and tracking, and information retrieval, and has a demonstrated knack for compact, high-performing models (e.g., AnoNet’s 65k-parameter architecture achieving SOTA anomaly detection). Comfortable across PyTorch, TensorFlow, HuggingFace and production languages like Python and Rust, he also experiments with generative AI, LLMs/VLMs and vector databases to bridge research and applied products. His background combines a strong academic foundation (Master’s in Deep Learning and Computer Vision, University of Waterloo) with industry experience at Zefr and Fugro, and past work deploying automated redaction and road-crack detection systems. He publishes and communicates ideas (Towards Data Science author), indicating an ability to translate technical advances for broader audiences. Based in Toronto, he’s focused on practical, efficient ML solutions that scale beyond the lab.
8 years of coding experience
6 years of employment as a software developer
Senior Secondary Certificate, Distinction, Senior Secondary Certificate, Distinction at St. Xavier's School and Junior CollegeHigher Secondary Examination, Computer Science, Distinction, Higher Secondary Examination, Computer Science, Distinction at Fr. Agnel Multipurpose School & Jr. CollegeBachelor of Engineering - BE, Electronics and Telecommunication, First Class with Distinction, Bachelor of Engineering - BE, Electronics and Telecommunication, First Class with Distinction at University of MumbaiMaster's degree, Deep Learning and Computer Vision, CGPA - 90.6/100, Master's degree, Deep Learning and Computer Vision, CGPA - 90.6/100 at University of Waterloo
Marathi, Hindi, Punjabi, English
Github Skills (113)
anomaly-detection10
python10
model-management10
prompt-engineering10
process-monitoring10
deep-learning10
gpu10
nvml10
mlmodel10
bounding-box10
mlflow10
audio10
nvidia-smi10
llmops10
ai10
Programming languages (5)
Github contributions (5)
msminhas93/FabricDefect
Apr 2019 - Mar 2020
A weakly annotated fabric defect dataset. Contains 24 512x512 images. Every image has a 512x512 mask. The mask has one rectangular bounding box covering the entire defect but has pixels of fabric without defect. This makes the defect detection task challenging.
Contributions:7 commits, 6 pushes, 1 branch in 11 months
bounding-box
msminhas93/CompactCNN
Apr 2019 - Dec 2020
PyTorch and Keras implementation of CompactCNN for Anomaly Detection in textured surfaces.
Contributions:11 commits, 9 pushes, 1 branch in 1 year 8 months
anomaly-detectionkeraspytorch
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.