I have spent six years selling and supporting the tools that test the web.

Cypress · Replay.io · Checkly · now Resolve AI. Solutions engineering, then Head of Solutions for North America.

• I did not come to agents from research. I came from six years of watching deterministic tests fail in production.
• Built InterIntel in 2023 — one of the first agentic coding tools.
• Before all of it: founder, UX consultant, therapist, chef.

projectJanus

started2026-05-09

team size1

budget$100 / month

Every line on every slide in this deck is editable — click it and type.

You don’t write tests. You write guides.

That is the last time this talk explains the product. Everything after it is evidence about the factory that built it.

One product. Three hosts, six processes, three languages, four test runners — one person.

Every line above is a seam two agents can disagree across, and not one of them fails loudly.

Throughput went up. The machinery required to keep it honest went up faster.

14 enforcement mechanisms, all built between Aug 2 and Aug 22. None in the first twelve weeks.

Every capability adds a failure. Every failure adds machinery. The overhead never comes back down.

"It is documented in agent memory … and being documented did not stop it." — scripts/tracker, line 25

The harness is small. It is also 84% bookkeeping.

90% of dispatches got a record. 209 changelog fragments got written. None were ever published.

A dispatch hands one issue to a subagent working in its own git worktree. A brief goes in; a handoff record — what it changed, what it verified — comes out. The worktree is then deleted, so the record is the only thing that survives.

A commit-msg hook blocks any commit naming a dispatched issue until its handoff exists. Writing a fragment is gated too — 209 written. Collating them into the changelog is gated by nothing — 0 published.

They had the rule in context. It governed them 7% of the time.

Same rule. Same repo. Same agents. Loaded both times — subagents receive every level of the CLAUDE.md hierarchy; only the built-in Explore and Plan agents skip it. The only variable was how close the rule sat to the prompt.

Your CLAUDE.md is not configuration. It is a suggestion with roughly 1-in-14 odds. The brief is the instrument that works.

None of this was falling behind. It was a decision I made on purpose.

A team pushes this to other people. Specialists, reviewers, a QA function, someone who owns the changelog. I had me.

An agent takes you as far as you can judge.
Not as far as it can write.

Vygotsky's zone of proximal development is the gap between what you can do alone and what you can do with help from someone who knows more than you. The scaffolding comes from above.

There is nobody above you here. An agent does not know more than you. It knows differently, and faster, and it cannot exceed your ability to check its work.

I was the only reviewer in the system, and the only component with no gate in front of it.

The line never comes back down.

One developer. One $100/month plan. 108 days. I did not write the code.

Questions.

Everything in this deck is a number out of one repo, and every one of them is in the talk notes if you want to argue with it.

jonathan canales · linkedin.com/in/modern-sapien