Senior Machine Learning Engineer, 3D Data

Overview

Build and scale the data infrastructure that powers Roblox's 3DGen foundation models, partnering with AI researchers to move from experimental datasets to high-fidelity 3D data synthesis and evaluation.

What you'll do

• Architect and maintain automated pipelines for ingestion, cleaning, and pre-processing of 3D datasets spanning petabytes of data.
• Research and implement synthetic data creation pipelines.
• Collaborate closely with ML Engineers and Data Scientists to understand their data requirements, build tooling that streamlines their data workflows, and troubleshoot data-related system issues.
• Build and own evaluation, automating both heuristic-based metrics and human-in-the-loop interfaces to evaluate and benchmark training datasets and in-house foundation models.
• Design and optimize high-throughput, low-latency Inference APIs for internal and external consumer access.
• Actively participate in literature reviews and paper reading groups to identify and implement the latest optimizations in generative modeling.
• Implement monitoring of pipeline health and optimize data loading to ensure GPUs are used efficiently.

What you'll need

• 5+ years of experience as a research-focused data systems engineer (preferably working with 3D generative models).
• Expertise in building scalable ML data pipelines for both batch and real-time environments.
• Experience working with and processing very large datasets (Petabytes or more).
• Versatile generalist, comfortable with several languages and technologies and adaptable in any situation.
• Team player and technical leader who actively mentors peers, drives technical excellence, and takes ownership of leading and delivering key features and projects across team boundaries.
• Python proficiency: can write high-quality Python code for automation, tooling, and infrastructure management.
• Passion for the potential of generative AI, particularly in creative domains like 3D/4D content.
• A Bachelor's degree or equivalent experience in Computer Science, Computer Engineering, or a similar technical field.

Nice to have

• MLOps and data versioning expertise: experience with experiment tracking platforms (e.g., Weights & Biases, MLflow) and managing versioning for massive-scale datasets.
• Custom tooling and annotation infrastructure: proven track record of developing internal human-in-the-loop workflows and annotation tooling tailored to 3D or video data.
• Low-level optimization (C++): strong C++ proficiency to optimize data-loader performance and extend underlying core engine functionality.
• Game engines and DCC tools: hands-on experience developing in Roblox Studio or working with digital content creation tools such as Blender, Unreal Engine, or Unity.

Details

• Location: San Mateo, CA headquarters.
• Onsite Tuesday, Wednesday, and Thursday, with optional presence on Monday and Friday.
• Annual base salary range: $212,490 to $295,250 USD; actual pay depends on job-related factors. Full-time employees are also eligible for equity compensation and benefits.
• For US-based roles only: the Company may not be able to employ candidates who have United States work authorization related to certain U.S. visa categories, or support future H-1B sponsorship at this time.