Roblox - San Mateo, CA

posted about 1 month ago

Full-time - Senior
San Mateo, CA
Professional, Scientific, and Technical Services

About the position

At Roblox, we are on a mission to connect a billion people with optimism and civility through immersive digital experiences. As a Senior / Principal Machine Learning Platform Engineer, you will play a crucial role in building the next generation of Machine Learning (ML) Ecosystem Tooling. This position is pivotal in shaping the future of human interaction by solving unique technical challenges at scale and creating safer, more civil shared experiences for everyone. You will have the opportunity to impact the Roblox platform and the industry by enabling our developers and creators to transition from an ML idea to production in weeks or less. We are looking for accomplished ML and Systems Engineers who are eager to build large-scale ML systems from the ground up. In this role, you will be responsible for developing service APIs for various ML Infrastructure components, including the Serving Layer, Metadata Store, Model Registry, Feature Store, and Pipeline Orchestrator. You will collaborate directly with partner teams to identify opportunities for model optimization and work across organizations to build tooling, interfaces, and visualizations that enhance the ML experience at Roblox. Your contributions will help elevate the state of the Roblox platform and the industry by advancing managed end-to-end ML Pipeline Development and Automation. You will also be part of a team that handles thousands of model experiments daily, supporting various functions such as ranking, recommendations, content moderation, fraud prevention, and studio creative tooling.

Responsibilities

  • Develop service APIs for ML Infrastructure components including Serving Layer, Metadata Store, Model Registry, Feature Store, and Pipeline Orchestrator.
  • Work directly with partner teams to identify opportunities for model optimization.
  • Collaborate across organizations to build tooling, interfaces, and visualizations that enhance the ML experience at Roblox.
  • Impact the state of the Roblox platform and the industry by advancing managed end-to-end ML Pipeline Development and Automation.
  • Enable developers and creators to transition from an ML idea to production in weeks or less.
  • Support the handling of thousands of model experiments per day for various functions such as ranking, recommendations, content moderation, fraud prevention, and studio creative tooling.

Requirements

  • 3+ years of professional experience working with scalable, distributed systems.
  • Well versed with the Model Development Lifecycle from initial ad hoc analysis in notebooks to monitored services in production and back again.
  • Experience deploying and maintaining an ML model in production, ideally at scale.
  • Experience developing ML Platform features that are user-friendly for MLEs and Data Scientists.
  • Ideally have experience with GPUs, including reading GPU profiles, debugging CUDA kernels, and optimizing GPU workloads.
  • Ideally have experience with large-scale distributed training across hundreds of GPUs for multiple weeks.
  • Understand best practices around Data and Model management.
  • Ideally understand the differences between unstructured and structured data and how to optimize data loading into training jobs.
  • Bachelor's degree in Computer Science, Computer Engineering, Data Science, or a similar technical field.

Nice-to-haves

  • Experience with cloud-based ML services and platforms.
  • Familiarity with containerization technologies such as Docker and Kubernetes.
  • Knowledge of data engineering practices and tools.
  • Experience with A/B testing and experimentation frameworks.

Benefits

  • Industry-leading compensation package
  • Excellent medical, dental, and vision coverage
  • A rewarding 401k program
  • Flexible vacation policy
  • Flexible and supportive work policy (Roflex)
  • Roblox Admin badge for your avatar
  • Free catered lunches five times a week
  • Several fully stocked kitchens with unlimited snacks
  • Onsite fitness center and fitness program credit
  • Annual CalTrain Go Pass
© 2024 Teal Labs, Inc
Privacy PolicyTerms of Service