Nvidia Deep Learning Engineer Generative AI 3D Reconstruction 

Japan, Tokyo 

460472388

What you will be doing:

• Implement and improve deep learning models to generate high-fidelity 3D representations from 2D images using the latest advances in neural rendering, Neural Radiance Fields (NeRF), Gaussian Splatting and photogrammetry.

• Design workarounds to eliminate artifacts that result from real-world adversarial effects, such as non-stationary scenes and lens irregularities.

• Optimize algorithms for efficient training and deployment of these models for the reproduction of very large scale real-life environments.

• Help customers to build digital twin solutions that combine the strengths of neural- and classic- rendering technologies for the generation of training and test data for computer vision foundation models.

What we need to see:

• University degree, or equivalent knowledge, in Computer Science, Computer Engineering, Electrical Engineering or Physics/Mathematics degree with Computer Science experience.

• Proficiency in C++, Python, data structures and algorithms and a solid understanding of computer architecture and operating systems.

• 5+ years experience developing algorithms in more than one of the following areas: virtual / augmented reality, photogrammetry, 3D reconstruction, NeRF / 3DGS / neural rendering, visual generative AI, foundation models, vision language models, style transfer.

• Strong understanding of traditional graphics rendering pipeline, such as rasterization and ray-tracing, OpenGL/Vulkan and shading languages.

• Knowledge of gaming engines, such as UE/Unity.

Ways to stand out from the crowd:

• Experience with multi-modal generative AI.

• Experience handling very large datasets and using large-scale cloud infra for training and deploying AI models.

• Contribution to open-source projects (can be your private project). Please provide link to github repository.

• Extensive experience in contributing to the field of visual generative models, with a strong publication record in top conferences or journals (e.g. NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, SIGGRAPH, etc.)

• Achievements in programming or machine learning competitions, such as Kaggle, HackerRank, TopCoder, etc.