MTS - Infrastructure

Infrastructure Team

Omnifold trains custom AI models that help planners forecast the future. We are hiring for our Infrastructure Team, who own the systems that make everything else possible.

What makes this job interesting:

We train a unique model for each customer, which means model training and inference work differently here than at any other company. You’ll never get more reps building model training infrastructure!

Our team has very fast iteration speed but needs robust monitoring to pick up signal on user patterns. This is especially important as our application interface for AI-driven forecasting is unique on the market.

What you’ll own

Deployment: Reliable processes for getting models and services into production

Security: Data isolation between customers, product security, infrastructure hardening (SOC2 compliance and beyond)

Cloud resource management: GPU allocation, instance sizing, cost optimization

Monitoring and logging: Visibility into what's running, what's failing, and why

Data and ML ops: ETL pipelines from varied customer data sources, model versioning and lifecycle management

Automated testing: Building the test infrastructure that lets us ship with confidence

What we’re looking for

Experience with cloud computing (especially GPU workloads), CI/CD infrastructure-as-code. We run on AWS

Familiarity with or interest in ML workflows

Security fundamentals: encryption, access controls, compliance basics

Python proficiency

Must have a strong Computer Science background

Location: San Francisco (in-person, 5 days per week)

Omnifold’s Mission

Every bad forecast has a physical consequence. Unnecessary goods are manufactured, shipped, and stored. Emergency air freight is needed for misallocated products. Poor production planning means workers show up with nothing to do, or work frantic overtime. Inefficiency is everywhere.

Our mission is to eliminate waste and accelerate growth for every company with physical products.