Hardik Dava
Computer Vision Engineer at Graswald AI
Hanover, Lower Saxony, Germany
Summary
🤩
Rockstar🎓
Top SchoolHardik Dava is a pragmatic Computer Vision Engineer with 4 years of experience building and deploying deep learning systems for 2D/3D perception, from neural radiance fields and diffusion models to edge-optimized object detection and segmentation. Based in Hanover, he blends strong research roots in stereo reconstruction and surgical AR with production experience shipping models to Jetson, EdgeTPU, OAK and other edge devices at companies like Graswald AI and corvitac. He is fluent across PyTorch/TensorFlow/ONNX toolchains, MLOps practices and full-stack APIs (FastAPI/Tornado), and has contributed improvements to high-profile open-source CV projects including supervision and YOLO variants. Notably, his background in laser and photonics informs a hands-on, physics-aware approach to 3D vision problems and synthetic data generation that improves model robustness in real-world deployments.
4 years of coding experience
3 years of employment as a software developer
Bachelor of Engineering - BE Mechanical Engineering, Bachelor of Engineering - BE Mechanical Engineering at Gujarat Technological University (GTU)Master's degree Laser and Photonics Technology, Master's degree Laser and Photonics Technology at Leibniz Universität Hannover
Gujarati, English, German, Hindi
Github Skills (9)
object-detection10
computer-vision10
pytorch10
machine-learning10
python10
image-processing10
coco9
tensorflow8
instance-segmentation8
Programming languages (9)
Github contributions (5)
roboflow/supervision
Apr 2023 - Dec 2024
We write your reusable computer vision tools. 💜
Role in this project:
ML Engineer
ML EngineerContributions:56 reviews, 45 PRs, 18 pushes in 1 year 7 months
Contributions summary:Hardik primarily contributed to the implementation and improvement of computer vision models within the `supervision` repository. Their work involved integrating YOLOv8 masks, adding support for COCO dataset import/export, and fixing various bugs and import errors within the object detection core. These changes demonstrate a focus on enhancing object detection functionality, improving usability, and supporting different model outputs, including YOLOv8 and related functionalities.
classificationcococomputer-visiondeep-learningimage-processing
hardikdava/gaussian-splatting-lightning
Jul 2024 - Dec 2024
A 3D Gaussian Splatting framework with various derived algorithms and an interactive web viewer
Contributions:6 PRs, 95 pushes, 25 branches in 4 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.