Seven people build all of it.
Every part of Core 137 — the takeoff canvas, the cost maths, the workspace around them and the invoice at the end — is designed, written and shipped by the people on this page.
That covers the whole width of the product: drawings and measurement, the cost database behind an estimate, scheduling, the file archive, billing and the account itself. One team carries all of it end to end, which is why the seams between those surfaces are the part we care most about.
It also covers the parts nobody advertises — the migrations, the audit trail, the release checks, the support thread at seven in the morning when a bid is due. There is no separate group here for the unglamorous half of the work.
Jose Baloa created the platform.
Founder and Chief Executive Officer. He built the first version of Core 137 and still owns its direction — what gets made, in what order, and how it has to behave on a jobsite before it is allowed to ship. He reads the support queue.
The rest of the table.
Six executives, one for each discipline the company actually needs. Every one of them builds; none of them exists to manage the others.
Small team, audited machinery. Seven people can only cover this much product because the boring parts are automated and checked. These are properties of the system, not intentions — each one is enforced on every change.
Every tenant table carries a row-level security policy, so a query that forgets its filter returns nothing instead of another company’s records. An audit command re-checks the whole schema for gaps.
The database is declared in files and rebuilt from them. It has been dropped and rebuilt from empty end to end, which is the only way to know the source is the whole truth.
Nightly, to a provider separate from the one running the database, and the restore path is exercised rather than assumed. A backup nobody has restored is a hope.
The host is generated from source and has been rebuilt from scratch on purpose. Nothing important lives only on a running machine, so recovery is a command rather than an archaeology project.
Types, an automated rulebook and the full test suite run before anything ships. The rules are written down and machine-checked, which is what keeps a small team consistent with itself.
TLS 1.3 in transit, encrypted at rest. Tenant isolation enforced by the database, not the application. Restore-tested backups, held with a separate provider. Key-based access only — no password or root login to production.
How we work.
Three commitments a seven-person company can actually keep. The table above is what makes them checkable rather than aspirational.
- 01.You reach the builderA support thread lands with the person who wrote the surface you are asking about. There is no tier one, because there is no tier two to escalate to.
- 02.Every surface has one ownerWhoever builds a surface owns its data model, its tests, its migrations and the support it generates. Nothing is somebody else’s problem by design.
- 03.A decision takes a conversationChanging direction is an afternoon, not a quarter. We treat that as the one real advantage of staying small, and we spend it deliberately.
Talk to the people who build it.
Seven people, one of whom wrote the surface you are asking about. Support threads reach them directly — there is no tier one.