Machine Learning Engineer (3D Gaussian Splatting & NeRF)

Summary

M-XR is a deep tech startup with a mission to make the 3D digital world look real; whether that be the graphics in a computer game, the CGI in a movie, or a product line photoshoot. We are building a solution that empowers 3D creators and enables the creation of productions at a speed, scale and quality not found anywhere else in the industry. Over the past three years we’ve developed foundational technology capable of capturing real world objects and accurately predicting their material properties, enabling the creation of ultra-realistic production-ready digital copies.

Curiosity and creativity are at the heart of M-XR. We feel strong that asking questions and looking at problems from new perspectives across departments is key to pushing the envelope for what is possible! We are looking for skilled individuals who share this passionate curiosity, question the norm, and have the willingness to explore something brand new. If you are an engineer or developer that shares this passion about shaping the future of 3D we would love to hear from you.

Description of work to be performed

As a Machine Learning Engineer at M-XR specializing in Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS), you will play a pivotal role in advancing our capabilities. Your focus will be on implementing cutting-edge computer vision algorithms for NeRF/3DGS and exploring ways to enhance these technologies by integrating segmentation, language embeddings (e.g., CLIP), and other advancements in 3D machine learning.

Leveraging M-XR’s proprietary dataset—the highest-quality ultra-realistic 3D dataset of its kind—you’ll tackle innovative projects that push the boundaries of what’s possible in machine learning and 3D. Key initiatives include 3D asset relighting and extracting ultra-realistic material properties directly from NeRFs/3DGS. A core part of your role will involve adapting, implementing, and enhancing open-source research and models such as SAM, Stable Diffusion, CLIP, DINO, NeRF, and 3DGS to create solutions tailored to our unique use cases.

Your contributions will directly set new standards for realism and quality in 3D content creation, with applications in major film and game productions. This role offers an exciting opportunity to tackle complex challenges, develop groundbreaking technologies, and witness the tangible impact of your work on the future of the entertainment industry.

Ideal Candidate

• Creative and innovative thinker

• Resourceful and effective problem-solver

• Clear, articulate, and proactive communicator

• Strong user-focused mindset

• Collaborative and supportive team player

Requirements

• Proficiency with a major industry ML framework (e.g., PyTorch, JAX, TensorFlow)

• Expertise in writing production-quality Python code

• Experience with CUDA programming

• Familiarity with Git and version control systems

• Strong understanding of computer graphics principles

• Hands-on experience with 3D Gaussian Splatting, either through contributions to open-source NeRF/3DGS repositories or solving relevant problems in the field

• Practical experience in training and fine-tuning diffusion models (a plus)

• Knowledge of C++ (an advantage)

Please ensure your CV is attached.

Best,

M-XR