Welcome

This documentation describes how the Safua platform works: the residency structure, how reviews are conducted, how scores are computed, and the interfaces available to learners and teams.

Where to start#

• New to Safua. Read How Safua Works for an end-to-end walkthrough of the residency: the three phases, what a mission is, who reviews it, and what a learner graduates with.
• Evaluating for a team. See Enterprise for team accounts, observability, and procurement, and Pricing for plan comparisons.
• Learner already in a residency. Tactical guidance (submission flow, how a specific score was produced, Proof Profile assembly) is documented per surface as features ship. This section covers the conceptual model that every surface rests on.

In this section#

• How Safua Works — the platform end to end, from Foundation calibration through to a public Proof Profile.
• The Residency Model — why Safua is structured as Foundation → Build → Prove, and how that differs from bootcamps, MOOCs, and degrees.
• The Five-Dimension Rubric — the five dimensions reviewed work is graded against, with examples of strong and weak scores.
• Confidence Score — how the single numeric readiness signal is computed and what specific values (0.93, 0.60) mean in terms of mission performance.

Conventions in this documentation#

• Technical mechanisms are described, not promoted. Where the product makes a specific claim (for example, "Confidence below 0.60 triggers an additional review pass"), the rule is stated verbatim and the math is available to audit.
• Where a feature is in progress, it is labeled "In progress" rather than described as if it already ships.
• If something in these docs doesn't match what you see on the platform, that is a bug. Report it to support@darkolab.com and it will be reconciled in code or in copy.