Machine Learning Engineer – 3D Gaussian Splatting & Neural Rendering
Description
Lit8 develops advanced AI systems for photorealistic visual content generation. In this role, you will develop, train, and optimize 3D Gaussian Splatting (3DGS) and neural rendering methods to reconstruct high-fidelity 3D scenes from images.
We are looking for someone with strong hands-on experience in 3DGS, neural scene representations, and differentiable rendering, capable of achieving photorealistic results while improving reconstruction quality, efficiency, and rendering performance.
Minimum Qualifications
- Master’s degree or Ph.D. in Computer Science, Computer Graphics, Computer Vision, Machine Learning, or a closely related field.
- At least 2 years of hands-on experience in 3D reconstruction, neural rendering, or related computer graphics research, including practical experience with 3D Gaussian Splatting.
- Strong understanding of 3DGS representations, Gaussian optimization, densification, pruning, and rendering pipelines.
- Experience training, fine-tuning, debugging, and optimizing 3DGS-based reconstruction systems.
- Practical knowledge of multi-view geometry, camera calibration, camera pose estimation, and scene reconstruction.
- Experience improving reconstruction quality, visual fidelity, geometric consistency, and rendering efficiency.
- Strong programming skills in Python and PyTorch, with experience in GPU programming or differentiable rendering frameworks.
- Strong mathematical background in computer graphics, optimization, and 3D geometry.
Preferred Qualifications
- Ph.D. or research experience in computer graphics, neural rendering, or 3D reconstruction.
- Experience with advanced 3DGS architectures, including scalable, high-resolution, or dynamic scene reconstruction.
- Experience with NeRFs, neural scene representations, or hybrid reconstruction approaches.
- Knowledge of physically based rendering (PBR), ray tracing, path tracing, and inverse rendering.
- Experience with Mitsuba, Dr.Jit, nvdiffrast, or related rendering frameworks.
- Experience optimizing GPU kernels, rendering performance, and memory efficiency.
- Publications at SIGGRAPH, CVPR, ICCV, ECCV, or contributions to relevant open-source projects.
Key Responsibilities
- Develop, train, and optimize 3D Gaussian Splatting models for photorealistic scene reconstruction.
- Improve reconstruction fidelity, geometric consistency, surface detail, and rendering performance.
- Explore neural rendering and physically based rendering techniques to improve materials, lighting, and visual realism.
- Optimize training and inference pipelines for efficiency, scalability, and production use.
- Collaborate with research and engineering teams to integrate 3DGS methods into production applications.
What We Offer
- The opportunity to work on advanced 3D Gaussian Splatting and neural rendering research with real product impact.
- A research-driven environment focused on technical excellence and ownership.
- Attractive salary.
If you are passionate about 3D Gaussian Splatting, neural rendering, and building photorealistic 3D reconstruction systems, we'd love to hear from you.
Skills
Sign in to read and join the discussion.