Boutique Systems integration for HPC & AI

Bringing science and craft into HPC/AI/ML systems integration.

Fluid Numerics helps teams design, build, and operate high-performance systems—from GPU clusters and Slurm stacks to reproducible ML pipelines—without losing the details that make them reliable.

Cloud • Bare metal • Containers • ROCm & CUDA • Slurm

Ecosystem

Compute providers we work with

Google CloudCirrascale

Tools we build with

PackerTerraformMAASWarewulf

Technologies we integrate

SlurmGoogle WorkspaceTailscaleDatadogGrafana

Languages we speak

AMD ROCmNVIDIA CUDAMPI & OpenMPFortranPythonC/C++

Capabilities

Thoughtful systems work. Built by practitioners who care about reliability, performance, and the long tail of operations.

Cluster bring-up, done right

Bare metal or cloud with pre-baked and tested images that include schedulers and monitoring with repeatability in mind.

Researcher support

Open communication, office hours, tailored tutorials focused on designing workflows and optimizing research software.

Slurm & workflow plumbing

Partitions, accounting, QoS, preemption, and job ergonomics—so your users ship work instead of wrestling config.

Reproducible software stacks

Versioned containers and modules, curated base images, and CI that proves the stack works before users do.

Performance as a practice

Profiling, tuning, and scaling guidance using microbenchmarks, mini-apps, and your applications—methodical, measured, and documented.

Flexible management

Co-managed or fully-managed offerings to fill gaps in cluster management, administration, and user support.

Yes, we do that too.

A team with backgrounds spanning Applied Mathematics, Geophysical Fluid Dynamics, Scientific Computing, Computational Physics, Electronics, and Music is bound to have side projects.

Let’s build something solid.

Tell us what you’re running (or trying to run). We’ll help you design a system that’s maintainable, measurable, and built to last.