Senior Machine Learning Infrastructure Engineer

Roblox • San Mateo, CA

4d • $243,290 - $295,250 • Onsite

About The Position

Roblox is hiring a Senior Machine Learning Infrastructure Engineer to build scalable, reliable, and high-performance infrastructure for machine learning systems across the organization. The role supports matchmaking, a critical part of the player experience that assigns players to game servers and influences player groups, network latency, and session quality. The infrastructure operates at the scale of tens of billions of scorings and matching hundreds of millions of users to millions of games.

The engineer will work at the intersection of model architecture and ML infrastructure, partnering with modelers, data and AI infrastructure engineers, and product teams. They will design and own large-scale data engineering and streaming systems, build online and continual learning systems, investigate end-to-end model, data, and systems tradeoffs, and establish best practices for scalable, reliable, and cost-effective ML infrastructure. They will also collaborate with data scientists, ML engineers, platform teams, and product stakeholders to build and operate robust ML platforms, while keeping abreast of developments in machine learning and infrastructure.

Requirements

• At least 5 years of experience designing, building, and deploying large-scale machine learning and data systems in production environments.
• At least 3 years of experience tech-leading ML infrastructure or data engineering teams.
• Hands-on experience with large-scale streaming and data processing systems such as Kafka, Flink, Spark, or Beam, and real-time feature pipelines.
• Strong communication skills and a collaborative approach to problem solving.
• BS, MS, Ph.D. in Computer Science or Engineering, or equivalent experience.

Nice To Haves

• Experience building or operating online, incremental, or continual learning systems for recommendation, ranking, or search.
• Deep production experience with low-latency key-value stores and feature registries such as Redis, Feast, Tecton, or Cassandra.

Responsibilities

• Partner with modelers, data and AI infrastructure engineers, and product teams to advance large-scale model training and serving.
• Design and own ML data engineering and streaming systems, including real-time event ingestion, streaming feature computation, training-data generation, and data quality and freshness guarantees.
• Build low-latency data pipelines using technologies such as Kafka, Flink, Spark Structured Streaming, and Ray, while maintaining consistency between online and offline features.
• Build online and continual learning systems that adapt models to user behavior and content shifts in near real time.
• Implement incremental and streaming training, online model updates and fast model refresh, embedding and parameter-server synchronization, and delayed-feedback and label-joining pipelines.
• Support safe model deployment through monitoring, drift detection, and automated rollback.
• Investigate model, data, and systems tradeoffs across streaming pipelines, distributed training, low-latency inference, and production serving.
• Design efficient cache strategies, apply pruning and quantization, optimize GPU utilization and memory efficiency, and develop custom kernels when needed.
• Own ML infrastructure architecture and establish best practices for scalability, reliability, and cost-effectiveness across training, serving, and feature infrastructure.
• Work with data scientists, ML engineers, platform teams, and product stakeholders to design, implement, and operate robust ML platforms.
• Stay current with machine learning and infrastructure trends and support adoption of leading-edge technologies and practices.

Benefits

• Equity compensation for full-time employees.
• Employee benefits are referenced on the company benefits page; specific benefits are not listed in the posting.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior