Open to opportunities

Shyam Sunder

@shyamsunder1

Message

Computer Vision Engineer specializing in deep learning and sensor fusion.

India

Message

What I'm looking for

I am looking for opportunities that foster innovation, collaboration, and growth in AI and computer vision.

I am a Computer Vision Engineer with a strong foundation in classical and deep learning techniques, particularly in 2D and 3D computer vision and LiDAR-camera sensor fusion. My journey began with a background in physics, where I specialized in cosmology and astronomy, which has equipped me with a unique analytical perspective in tackling complex problems.

During my internships, I developed a range of innovative solutions, including a Python extraction pipeline for tracking drywall sample parameters and an advanced YOLO model for detecting oversanding in robotic applications. My work has consistently focused on enhancing performance and accuracy, achieving impressive results such as a test IoU of 0.97 and a mAP of 0.94. I am passionate about leveraging my skills in machine learning and computer vision to create impactful solutions that drive efficiency and quality.

Experience

Work history, roles, and key accomplishments

Computer Vision Intern

10xConstruction.ai

Feb 2025 - May 2025 (3 months)

Developed a Python extraction pipeline and AWS S3 uploader to track spraying/sanding parameters, and extended Butterfly Viewer for rapid batch QA of robot performance. Engineered an OpenCV + Albumentations preprocessing stack and trained a SegFormer model, achieving high test IoUs. Led extensive YOLO model experimentation with RayTunes-driven hyperparameter sweeps and custom augmentations to detec

Python AWS S3 OpenCV Albumentations SegFormer YOLO RayTunes

Computer Vision Intern

Collaborative Intelligence

Nov 2024 - Jan 2025 (2 months)

Developed a dilapidated-document cropping system by training YOLO11 on 10,000+ annotated images, achieving a mAP = 0.95, and expanded predicted bounding boxes to improve cropping accuracy by 15%. Integrated SAM2.1 for finer segmentation, achieving 98% accuracy, and developed a robust post-processing pipeline using OpenCV for edge smoothness. Boosted processing speed by 60x through GPU-based data p

OpenCV Segmentation Document Cropping

Education

Degrees, certifications, and relevant coursework

Liverpool John Moores University

M.Sc. Data Science, Data Science

Currently pursuing a Master of Science in Data Science, with a specialized focus on Deep Learning and Computer Vision. This program enhances skills in advanced data analysis and machine learning applications.

Heidelberg University

M.Sc. Physics, Physics

Activities and societies: Thesis: A contribution to the simulation of the breaking of statistical isotropy effects in SU(2) CMB; Supervisor: Dr. Ralf Hofmann, ITP, Heidelberg University, Germany.

Completed a Master of Science in Physics, specializing in Cosmology and Extragalactic Astronomy. Thesis research involved simulating the breaking of statistical isotropy effects in SU(2) CMB.

Shiv Nadar University

B.Sc. (Research) Physics, Physics

Activities and societies: Minor: Mathematics; Thesis: Entropic Law of Gravitation and MOND corrections; Supervisor: Prof. Bijan Kumar Bagchi, SNU, India.

Obtained a Bachelor of Science (Research) in Physics, complemented by a minor in Mathematics. Thesis work investigated the Entropic Law of Gravitation and MOND corrections.