Varun Deshpande
Machine Learning Engineer · Computer Vision
Professional Summary
Machine learning engineer with 4 years in computer vision for manufacturing and retail. Builds detection models that run on edge devices on the shop floor, from data labelling to deployment.
Work Experience
Monsoon Grid Robotics – Machine Learning Engineer
Pune
- Trained a YOLOv8 defect detector on 85,000 labelled images, catching 96% of surface defects at 1.5% false alarms.
- Quantised and deployed the model on NVIDIA Jetson with TensorRT, reaching 42 frames/second on the line.
- Built an active-learning loop that cut labelling effort by 40% for each new product variant.
- Reduced manual inspection staff on 3 lines from 6 to 2 people per shift.
Shelfsight Analytics – Computer Vision Engineer
Mumbai
- Built a shelf-gap detection model for 220 grocery stores using images from store-staff phones.
- Wrote the data pipeline that turned raw photos into training sets in under a day.
Projects
Low-light PCB inspection – M.Tech thesis
- Image enhancement before detection raised recall on dim lines from 81% to 93%.
Download your resume to print it.
Crest
ATS-safeTwo-column layout with a small accent tab beside a bold name, accent rules between sections, and a slim sidebar for contact, skills and education. Crisp and easy to scan.
Two-columnEasy to scan
- No photo — text only.
- Page size: A4.
- Switching to another template later keeps all your content.