Shubham Patel
3D Computer Vision & Generative AI Engineer
Image-to-3D Generative AI • Neural Rendering • Photogrammetry • Multimodal AI Models
About Me
Senior AI Engineer specializing in 3D Computer Vision and Generative AI, with deep expertise in high‑fidelity 3D reconstruction, neural rendering, and multimodal Image/Text‑to‑3D AI generation.
Holder of two MSc degrees in Artificial Intelligence and Computer Vision. Proven experience designing, training, deploying, and industrializing AI systems from scratch as well as fine‑tuning large AI foundation models for production use cases.
Strong emphasis on data legality, robustness, and scalable synthetic data generation pipelines using physically‑based simulation environments for AI training.
Technical Skills
Programming Languages
Deep Learning & AI
3D Vision & Graphics
DevOps & MLOps
Cloud & Storage
Libraries
Professional Experience
Deep Learning Specialist (Computer Vision)
Dassault Systèmes
Sept. 2022 - Present
Full-timeAix-en-Provence, France
Leading 3D reconstruction projects for digital twin applications. Developing Image/Text-to-3D Generative AI pipelines and deploying end-to-end AI systems.
Computer Vision Engineer
Dassault Systèmes
March 2022 - Sept. 2022
InternshipAix-en-Provence, France
Photogrammetry R&D and Computer Vision applications for 3D reconstruction. Advanced AI reconstruction algorithms development.
Developer
DXOMARK
Feb. 2021 - Jan. 2022
Study ProjectBoulogne-Billancourt, France
Image Processing development. Built QT interfaces and implemented OpenGL shaders.
Back End Developer
SNCF Réseau
Nov. 2020 - Jan. 2021
InternshipSaint-Denis, France
Created Django REST API and documented endpoints using OpenAPI/Swagger.
Key Projects
Large-Scale Outdoor Scene Reconstruction
Reconstructed extensive outdoor environments using SfM/MVS pipelines for digital twin applications.
Indoor Scene & Object Reconstruction
Precise indoor reconstructions with complex lighting using neural rendering and Gaussian Splatting.
Multimodal Generative AI (Image/Text-to-3D)
Generative AI pipelines converting images or text into consistent 3D representations.
Robust Background Removal for Object Reconstruction
Deep learning segmentation model for isolating objects from cluttered backgrounds.
GPU Ray Tracing & Rendering
GPU-accelerated ray tracer with physically based rendering principles.
Synthetic Data Generation for AI Training
Automated pipelines using Blender and UE5 for generating training data.
Education
Computer Engineering
EPITA – École d'Ingénieurs en Informatique
2019 – 2022
Computer Vision, Machine Learning, Image Synthesis
Master 2 – Ingénierie Informatique
Sorbonne Université
2021 – 2022
Image Processing, Computer Vision, ML, Image Synthesis
Languages
Get In Touch
Interested in discussing opportunities, collaborations, or projects in 3D Computer Vision and Generative AI? Feel free to reach out.
Location
Aix-en-Provence, France
Currently at Dassault Systèmes