The example projects are evidence of a way of working. When an AI writes and maintains software, the good patterns change. These are the ones that show up again and again across the projects, each grounded in real code.
Every claim below points at something concrete in the repositories.
Across 22 active project boards, as of October 2026. FeatureBoard — the platform that builds the rest — is itself the biggest example of every paradigm below.
Every time data crosses a boundary — a file loaded, a surface parsed, a mesh repaired — the code says out loud where it came from, what it tried to do, whether it worked, and if not, why. When an AI is the one debugging, dense, tagged, structured logs are worth more than clever code. You can always turn verbosity down; you can't recover a reason that was never printed.
The tessellator narrates every face: [STEP-EDGE], [REPAIR], [SNAP-WELD], [PLANE-TESS]. A single run emits a full provenance trail of the geometry decisions.
When something isn't in the right shape, the code does X instead of printing "please run X and come back." If an AI response comes back badly framed, the correction lives inside the pipeline and re-runs automatically. If a mesh isn't watertight, it gets repaired before the solver ever sees it. You get the outcome.
InfinitusPie routes review-response generation through OpenClaw and auto-fetches three drafts the moment a review lands, with no "click to generate." STL Wind Tunnel runs STLRepair as a pipeline stage, so the mesh is watertight before you have to think about it.
Heavy work runs as explicit pipelines you can inspect and re-run; light-but-private work runs at the edge, on the user's own device. The split is deliberate: train and process on the workstation, infer where the data already lives.
Gaussian-splat training and the GPU video engine are staged pipelines (COLMAP/nerfstudio, FFmpeg/NVENC, YOLO). SlopRadar sits at the other end: it runs Gemini Nano in the browser, so the feed never leaves the device.
A surprising number of "hard" features become easy once you reframe them as: give the model context, ask for a judgement, get structured JSON back. The engineering moves into the framing and the schema, and the pile of heuristics goes away.
MarriageBank's "AI Banker" takes a plain-English activity ("back rub") and returns a scored deposit/penalty — a relationship-effort ledger built entirely on model judgement. SlopRadar hands a post to Gemini Nano and gets back a JSON slop/not-slop classification. Same shape, very different domains.
When a machine writes the code, "does it look right?" is the wrong question. Every feature and every bug fix ships with an executable check that proves the behaviour. The test suite is the review.
5,140 executable test files across the projects, most of them one per feature (test_*f-*.js) and per bug (test_*b-*.js), traceable to the item that motivated it. NativeCAD: 3,494 · FeatureBoard: 1,148 · InfinitusPie: 102 · STLWindTunnel: 95 · Letter Launcher: 94 · PointCloud: 40 …
FeatureBoard also runs a nightly suite — named checks like "Project Dashboard Load Test" and "API Health" — with 295 nightly test reports on disk. Correctness gets proven on a schedule.
Features start as structured metadata, before anyone opens an editor. A living featurelist.md, buglist.md, a project MAP.md, and an agent_work_log.md give the AI durable memory: what's built, what's broken, what changed, at what cost. That metadata is what lets features be planned, tracked and shipped automatically.
FeatureBoard itself (the board and its MCP server) tracks 1,064 shipped features and 1,255 fixed bugs as structured entries in 3.5 MB of featurelist.md + buglist.md, backed by 1 MB of agent_work_log.md (2,581 entries) and 1.8 MB of behavioural analytics. Across every board it is 5,007 features, 3,523 bug fixes and 15,167 work-log entries. Each work-log line carries the task ID, files touched, adds/deletes, and even token cost — e.g. Task: MBF-22 · Files: marriagebank.{html,css,js} · Add: 199 Del: 100. Plans in, provenance out.
Where most projects would npm install someone else's engine and glue it together, these implement the hard part themselves. Owning the core means it's understood, debuggable, and bendable to the exact problem. The imports cover the boring parts; everything interesting is hand-written.
Gaussian Splats is our own WebGL splat renderer: the PLY parser, the spherical-harmonic evaluation, and the view-sorted gaussian-quad rasterizer are written from scratch rather than borrowed from gsplat.js or a ported viewer. NativeCAD hand-writes its whole solid modeling kernel (STEP (ISO 10303) parser, surface tessellator, and B-rep modeling ops up to variable-width edge blends) instead of pulling in opencascade.js. The Wind Tunnel is a from-scratch flow solver, volume mesher, and free-surface water model, with no OpenFOAM underneath. The Point Cloud Visualizer runs a custom wavelength-propagation engine. Comet Rex has its own THREE-free world model. And the Rust Installer takes it to the extreme: standard library only, zero crates.