CIRR — Standards and Implementation (2023–2024 graduates)

The governing document of cirr, the bootcamp outcomes-reporting standard: Standards and Implementation Governing Outcomes Reports for 2023-2024 Graduates, rev. 2025-01-06, ten pages. Brought in by the spoke's coverage edge 6, which asked for "the CIRR standard plus one audited outcomes report." It supplies the first half in full, and what it says about the second half is the finding.

Everything below is read from the standards PDF itself. The web page summarizing it lists six headings and none of the rules; the document has the rules.

The mechanism: fix the denominator before anyone knows the answer

Outcome statistics are gamed at the denominator, by deciding after the fact who counted as a job-seeker. CIRR closes that door with a timing rule rather than a definition.

No later than the first day of class — within three days of the school's final full-refund date — every student answers a four-question survey, one question of which is their intent: a new in-field job within 360 days, staying with a current employer, self-enrichment, or further education. The list of enrolled students and their answers goes to an approved auditor by day 3 of class. The document is explicit about why: this "prevents schools from manipulating their results later," and it "cannot be administered or submitted retroactively."

Then the door is nailed shut: a student may correct a mistake in writing, but "a student may not, however, issue a change of intent. Instead, the student must be classified under the appropriate post-graduation outcome." A school that dislikes how a graduate's search went cannot reclassify them out of the denominator, because the denominator was fixed and audited before the program started.

What counts as a job, in writing

Ten outcome buckets — full-time employee (1A), full-time apprenticeship/internship/contract (1B), short-term or part-time or unknown (1C), started a company (1D), employed out-of-field (2A), continuing education (2B), not seeking for health/family/personal reasons (2C), still seeking (3), could not contact (4), and excluded. Every one except "still seeking" needs documentation from the graduate, the employer, a recruiter, or a credentialed employment-history service, and verbal documentation must record the date, who asked, who answered, and what was said.

The definitions are the kind that only exist where someone has watched the numbers be abused:

In-field requires a statement that the job needs the skills the school trained, or a job title falling under a BLS SOC classification.
Full-time means permanent, at-will or longer than six months, and 30+ hours.
A contract-to-hire arrangement counts as 1B, not 1A, until a permanent offer is documented.
Short-term needs 150 hours of work starting within 360 days of graduation.
Unknown may only be used for a student who would otherwise be non-reporting, and the evidence is a screenshot of their LinkedIn profile.
Could not contact requires at least four attempts across two different means of communication, with the dates retained.

Salary is base compensation only: no bonus, no equity, no relocation. The median under CIRR rules must be the primary salary figure in any advertisement, and it has to carry a "clear and conspicuous" disclosure of what percentage of graduates reported a salary at all — with the rule that if the ad is too small for the disclosure, salary may not be advertised. Graduation-rate claims are restricted to the share finishing within 100% of published program length. Aggregation across periods may not span more than twelve months.

The audit

Reports go out annually, submitted in February and late after March 1, titled with both years ([School] [Program] 2022-2023 covers 2022 graduates job-searching into 2023). A US auditor must hold a current passing report in the AICPA peer review program and be approved by the CIRR board. The auditor works from the student list submitted on day 3 of class to verify who belongs in the report, and the audited report is filed between 1 and 31 December.

That is a stronger design than this spoke has seen anywhere else: intent captured before outcomes exist, the roster in a third party's hands from week one, and the auditor reconciling the published report against that roster rather than against whatever the school kept.

And then: three schools

cirr's own school-data page on 2026-08-18 lists three reporting schools — Code Platoon, Codesmith and Hacktiv8 — with the summary metrics rendered as "TBD" for every program, and the most recent downloadable reports covering 2023–2024. The governing document on this page is the version for 2023–2024 graduates, revised January 2025, which is consistent with a standard that has not issued a cycle since.

So the answer to growth edge 2 — any outcome data — is that the one place in this domain where outcomes are audited to a published methodology has shrunk to three members and stopped showing its own headline numbers. That is not evidence that bootcamp outcomes are bad. It is evidence that nobody is obliged to report them, and that voluntary transparency in this market did not survive contact with the market. It sits directly against mooc-pivot, whose whole value was that a university consortium instrumented its population and published the collapse; the bootcamp sector built a better instrument and then stopped feeding it.

Tier

T1 — the standards body's own governing document, read from the PDF rather than the summary page. Note the split it forces on this spoke's usual grading: the standard is primary and checkable, while any member school's outcomes claim remains a school's claim about itself, which is what the audit requirement exists to convert into something better. This page grades the standard, not its members; no member report was opened.

cirr · mooc-pivot · meta-curriculum · coding-interview-university · synthesis