Role overview

UniversalAGI is building foundation models that learn physics directly from data, targeting engineering simulation—a category where AI has not yet fundamentally transformed the industry. The company works on real computational fluid dynamics and reservoir engineering problems for major industrial and defense organizations. As an ML Platform Engineer, you will design and operate the core execution platform that powers both internal research and customer deployments across data generation, simulation orchestration, training infrastructure, and production systems.

Key responsibilities

Build and maintain scalable infrastructure for data generation and simulation workflows, including job orchestration, scheduling, queues, retries, and observability systems.
Create reproducible pipelines for training, fine-tuning, and benchmarking with artifact management, version control, experiment tracking, and dataset lineage.
Own cost and performance tradeoffs across compute, storage, networking, and runtime efficiency to optimize platform operations.
Lead deployments of the company’s stack into customer environments on AWS, GCP, Azure, and hybrid cloud setups, handling secure networking, permissions, and data movement.
Build robust deployment patterns including environment provisioning, CI/CD, rollbacks, monitoring, and incident response protocols.
Partner directly with customers to ensure reliability, repeatability, and compliance under real-world constraints including security and data governance requirements.

Requirements

Strong software engineering foundation with demonstrated ability to write clean, reliable code and debug complex systems.
Hands-on experience building or operating infrastructure for ML and compute-heavy workflows, including pipelines, job orchestration, GPU compute, storage systems, CI/CD, and monitoring.
Mindset focused on measurable improvement and high standards for yourself and your deliverables.
Ability to make sound technical tradeoffs between quick fixes and robust solutions based on impact and customer needs.
Comfort owning work end-to-end with accountability for measurable outcomes.
Experience with workflow orchestration tools such as Ray, Kubernetes, or Slurm is beneficial.
Background with GPU infrastructure, distributed training systems, or simulation and HPC pipelines is a plus but not required.

What makes this role worth considering

You will work directly with the CEO and founding team to turn research into systems that power real customer deployments for some of the world’s largest industrial and defense organizations. The role offers significant scope and ownership in shaping the technical foundation of a rapidly growing AI company solving cutting-edge industrial problems with immediate real-world impact. The company maintains an in-person first culture with 5 days per week in the San Francisco office, daily covered meals, and comprehensive benefits including competitive equity, health coverage, wellness stipends, and immigration support.