Work
Assessment, deployment, techniques, research.
Engagements start with a small piece of fixed-fee work, so you have a number for your own system before committing to anything larger.
01 — Assessment
We take your workload and tell you what it could run at.
You give us a recording of an ordinary day on the real system. We run it on the hardware you use today, then run it again with our work in place, and report the difference.
The report is short: the number, the conditions it was taken under, and the specific changes in your system that would get you there.
Two to four weeks. Fixed fee, quoted before we start.
02 — Deployment
Our engineer works inside your team until it does.
One of our engineers joins the team that owns the system. They work in its codebase, sit in its reviews, and stay until the improvement is running in production against real traffic.
The work lands in your repository, in your language, covered by your tests. Once it ships you do not need us around to keep it running.
Every engagement has an end date and a fee agreed before it begins.
03 — Techniques
Our own methods, embedded in your system.
The method we use most often is called HST. It ships as a small library that links straight into the process already holding your data. There is no service to call and no container to run. Your data never leaves the machine it is on. Integration is a handful of function calls, and the result is exact: the same bits your current code produces.
It is free to use, including in production, and free to pass on. There is no key to install, nothing to renew, and no count kept anywhere. You can run it disconnected, in a plant, on an air-gapped network. What we charge for is the work of finding out whether it helps you and making it fit.
- Runs
- In‑process
- Result
- Exact
- Your data
- Stays put
- Integration
- No rewrite
Linked into your program. No service, no network.
Matches your existing computation, not an estimate.
Nothing is transmitted anywhere.
A small interface around what you already have.
04 — Research
We publish what we learn.
Knowing which systems this helps, and by how much, is most of what we are selling. It came out of running the experiments, so we publish the parts worth reading.
Start with the assessment.
Tell us what the system does, how often it recomputes, and how long that takes today. Estimates are fine at this stage.
Book a discovery call →