Build what actually matters
You ask: How do we build what matters?
AI can write code now, so the scarce part is deciding what is worth building.
AI can write a lot of code, so the hard part is choosing what to build. B Team is designed to start every piece of work from your goal and a clear done-when, check each change with a separate reviewer, and end with feedback from real users. You judge success by what gets used.
Most of this is planned. What exists today is the base: triage of new work, a coordinator, a pair of worker and reviewer, the gates, and a form that sets up a project.
How it answers
- You write your goal in your own words. The engine drafts one issue with a done-when, and you approve it. Planned, not built
- A planner shows which tasks can run together. You can change the plan. Planned, not built
- A separate reviewer checks every change, and it cannot change the work. Measured (pair runs)
- Only a person can accept a change. An agent cannot accept for you. Measured, with a recorded deviation
- Test users get a link and a few questions, with no account. Planned, not built
- Five measures show whether it was the right thing and whether anyone used it. Planned, not built
Reliability on this page
No agent is trusted, so every change is checked before you see it: a separate reviewer looks at it, and only you accept it (claims 10, 11 and 13). How we raise reliability
The honest limit
- Nothing measures yet how often a first result is accepted, or whether anyone used it. The five measures are meant to, and they are not built.
- We call the first full round "from your request to a finished improvement you approve". One goal becomes one accepted change. It is planned, and we have not set a date.
- You can try a change in a preview before you accept it. That is planned, it needs the sandbox first (issue #176), and it is not built.
For NGOs
Public value, impact for the people you serve, and accountability to your board and your funder. Once it is built, this is the promise we want to keep: "You can show your board or funder what was asked, what was built, and whether it was used." Today that sentence is not true. It is planned.
The technical detail
The claims behind this page, with their states as the evidence page records them.
| No. | Claim | State | What it serves | Check or issue |
|---|---|---|---|---|
| 17 | New work is triaged before anything starts. | Measured | Mailboxes, triage first (F4) | mailbox_checks (40); 1 real item |
| 34 | A coordinator plans a job. The machinery runs one step at a time. | Measured | Coordinator (#156) | coordinator_checks; #61 |
| 10 | A separate reviewer checks work before you see it. Nobody has shown that a reviewer catches a real fault. | Measured (pair runs) | Independent review (F13) | reviewed_checks; #28 |
| 11 | The reviewer cannot change the work. | Measured | Reviewer without write access (F13, S2) | reviewer_write_checks (5 of 5) |
| 13 | Nothing is published or merged without a person. No self-approval. The recorded deviation: whoever holds a project's key (Temporal token) to the engine's workflow service can send a decision under any name. | Measured, with a recorded deviation | Human approval (F9, S9) | inbox_checks; #32 |
| 33 | Assistants cannot answer a gate. | Built | Gates closed to assistants (F9, S9) | test_tools (5 tools withheld) |
| 29 | A page shows running or stopped, all jobs, waiting items, cost. | Built | Web interface (F17) | test_page; test_jobs_page |
| 31 | Set up a project from a form that opens a merge request. | Built | Project setup from a page (#239) | test_project_form (stand-in) |
| 45 | A planner proposes safe parallel work. You can change it. A register stops clashes. | Not built | Planner and register (F21) | #286 |
| 48 | A goal becomes an issue with a Done when, drafted by the engine and approved by you. | Not built | Goal to Done when (F23) | #287 |
| 49 | Test users answer a link and a few questions without any account. | Not built | Feedback and user tests (F24) | #288 |
| 52 | Five measures are recorded by the machinery and shown together. | Not built | The five measures (F27) | #290 |
The other four questions
-
Question 2
Choose your models, keep your freedom
You choose your models and where they run.
Read the answer -
Question 3
People and AI, building together
Agents write. People decide, guide and check.
Read the answer -
Question 4
Use AI and stay compliant
B Team produces the records your compliance needs as you work.
Read the answer -
Question 5
Fit your people and your budget
Work follows your rhythm. Much of it runs in batches.
Read the answer
Tell us you are interested
B Team is not offered to anyone yet. NGOs come first. If you want to hear when that changes, or you would like to be one of our first pilots, write to us.
Write to: [contact address to be decided]
This draft has no form and collects no data. Once there is an address, what happens next is meant to be this: we tell you whether we can work with your organisation yet. We promise no date and no price.