Senior ML Engineer - Foundation Data Pipelines for 3D/Video

Showcify; San Mateo (CA); On-site; USD 243,000 - 295,000; Full time

Job summary

Roblox is seeking a senior Machine Learning Engineer focused on data systems for its Foundation AI. You will architect and maintain high-scale data ingestion and preprocessing pipelines for multi-modal data spanning Petabytes, enabling research to production of 3D/4D content.

You will collaborate with researchers to scale data, design robust evaluation frameworks, and deploy high-throughput inference APIs while optimizing GPU utilization.

Qualifications

• Bachelor's degree or equivalent in CS/CE or a related field.
• 8+ years as a data systems engineer focused on ML/data pipelines.
• Experience building scalable ML data pipelines for large datasets.
• Proficiency in Python; strong C++ knowledge is a plus.
• Experience with cloud data platforms and distributed processing tools.

Responsibilities

• Architect and maintain high-scale data ingestion, cleaning, and preprocessing pipelines for multi-modal datasets.
• Generate synthetic data using image/video generation models.
• Bridge research and production: create training data pipelines for experiments.
• Build scalable evaluation frameworks with automated metrics and human-in-the-loop interfaces.
• Design high-throughput, low-latency inference APIs and optimize data loading for GPU workloads.

Skills

Python

C++

MLOps

Data pipelines

Distributed systems

Cloud platforms

3D/Video ML

Education

Bachelor's degree or equivalent in CS/CE or related field

Tools

Spark

Ray

Kubeflow

S3

Blender

Unreal Engine

Unity

Job description

Roblox is seeking a senior Machine Learning Engineer focused on data systems for its Foundation AI. You will architect and maintain high-scale data ingestion and preprocessing pipelines for multi-modal data spanning Petabytes, enabling research to production of 3D/4D content.

You will collaborate with researchers to scale data, design robust evaluation frameworks, and deploy high-throughput inference APIs while optimizing GPU utilization.