The software development platform
Skynet would have killed for.
Stable, scalable, and hands-free tools for building the next generation of software, intended to replace an obsolete stack not designed for the new world.
The Motor Way
Tibetan painters show the path to a calm mind as a climb: a monkey (distraction) drags an elephant (the mind) up a mountain. In the Motor Way, the elephant is your attention; the monkey is the work pulling it around: code, pull requests, merge conflicts, broken builds, and notifications.
As you climb, your agents take on more of that work. Habit still pulls you back to look at the code, even as it dawns on you that you don’t need to. The monkey falls behind, and the elephant walks calmly.
Start at the bottom to explore the Motor Way.
-
Monkey mind
Monkey mind
Your agents write code, but you still read it, review their changes, manage integration, and wait for builds and tests. You rely on tools designed around your participation at every step.
You are a serial being. The monkey leads the elephant, and you run after both of them.
-
One thousand arms
One thousand arms
You’ve begun to let agents handle multiple streams of work. Amp lets them combine concurrent changes, Gauss hosts their work without a human UI, and Piston runs their builds and tests remotely.
You still check their code. You’re used to knowing what’s happening by looking at it, and working without that familiar interface feels strange. You’re discovering how much of your job computers can already do.
-
Empty hands, clear mind
Empty hands, clear mind
You grow comfortable letting your agents handle the code. They write, review, test, and integrate changes without you following every step. You still question decisions and catch mistakes, but you no longer need to read the code to participate.
You stop measuring your contribution by how much of the work passes through your hands. You give direction and judge outcomes.
-
Calm abiding
Calm abiding
You do the new job: managing agents well. You plan with your teammates, give agents direction, and review the results. Your agents handle the code.
You provide productive resistance to endless agentic flow, like a rock redirecting water. You haven’t seen a line of code in ages.
The Motor Way is heavily opinionated. We know many people reject this approach today, and that’s okay. But if you’re open to working at a higher level of abstraction—letting agents handle the code and judging their work by its outcomes—there’s some real fun ahead.
Amp
Version control for massively parallel agentic work
Git represents versions as snapshots linked by commit ancestry. Amp represents versions as sets of patches with explicit dependencies.
-
Order-independent integration
The same set of patches produces the same code and conflicts, regardless of arrival order. Agents work concurrently without you managing integration order.
-
Resolve once and share it
A resolution is another patch. Agents exchange it with the conflicting changes, so each channel doesn’t need to solve the same conflict again.
-
Import from Git
Bring your existing repository into Amp, including branches, tags, and history. Keep the context behind your code as your agents move to a workflow built for parallel work.
Gauss
Repositories that land changes when the checks pass
GitHub centers code hosting on human participation. Gauss centers it on agents exchanging code under explicit channel policies.
-
Agents self-manage
Agents can submit changes, inspect failures, and push fixes through commands and APIs. Nobody needs to navigate a website or shepherd a merge.
-
Automatic checks and merges
All changes go through a merge queue designed to leverage Amp’s unique commutative properties, ensuring build retries are kept to an optimal minimum when automatic checks are in place.
-
No social proof
Stars measure attention. Approval tells you little when the human hasn’t inspected the code. Gauss bases integration on explicit channel policies and executable checks.
Piston
Builds, tests, and scripts, run on our machines
Generating code uses inference; building, testing, and running it needs compute too. Piston takes that work off your laptop. Agents send it the files they’re editing right now, or run a workflow from your repository.
-
Run changes without pushing
Agents can submit their current files, including unrecorded edits, instead of invoking a workflow against a repository ref. Of course, Piston tasks are designed to be used for CI as well, running as Gauss checks against merge candidates.
-
Fully isolated and fully informed
Piston runs builds, tests, scripts, and other tasks in isolated environments. Agents can easily retrieve structured logs, results, and output files to decide what to do next.
-
Agent work runs remotely
Fast remote execution keeps their work moving without using your laptop’s CPU and memory.