Introducing Temple Compute OS

For the last few months we've been heads-down building Temple Compute OS (tc-os), the managed platform that lets research teams define, run, and monitor computational workflows across HPC clusters and cloud environments without ever touching a scheduler config file.

Today we're opening early access.

Why we're building this

Every biotech, materials science, and AI research team we've talked to has the same problem: the science is the easy part. Getting a workflow to actually run (provisioning the right compute, wiring up the scheduler, babysitting long jobs, recovering from failures) eats weeks that should have gone to the research itself.

Temple Compute OS auto-provisions HPC or cloud compute for you, and gives you one interface to define, launch, and monitor workflows regardless of where they execute.

What's underneath

Temple Compute OS is a managed product, but it isn't a black box. It's built directly on top of horus-runtime, our open-source workflow execution engine. If you want to see exactly how workflows are scheduled, retried, and executed, or run the engine yourself outside of our managed platform, the code is public. Read more in our horus-runtime post or head straight to the repo.

Early access

We're rolling out Temple Compute OS gradually. Ask for beta access on the homepage and we'll reach out as seats open up. Early members get a launch discount as a thank-you for getting in before we're ready for everyone.