Product
Capable
Evidence-based skills assessment, mapped to SFIA 9.
An AI-powered skills assessment platform built on the SFIA 9 framework. Claude-driven STAR interviews adapt in real time to what a person actually says, producing evidence-based reports across four assessment modes. Neurodiversity-first by design and enterprise-ready.
Capabilities
Here is what Capable does.
We map every assessment to SFIA 9
Assessments map to the Skills Framework for the Information Age, so results speak the language your workforce plans, position descriptions and career pathways already use.
The interview adapts to what the person says
Claude-driven conversational assessment that follows up on what someone says, so a rehearsed answer to a fixed question list does not carry the interview.
You can run it at four different depths
Different depths and purposes, from lightweight self-assessment through to structured evidence-gathering for formal capability mapping.
Every rating cites the evidence behind it
Reports cite the responses that support each rating, so an assessment can be reviewed and argued with.
We designed it for people who interview badly
We designed this from the start for people who do not perform well in conventional interview formats, and we made that choice before the product existed rather than retrofitting it.
It is built for organisational rollout
Built for organisational rollout, with the administration and reporting a workforce programme requires.
The problem it solves
Most organisations do not know what skills they have. They have position descriptions, which describe what people were hired to do, and performance reviews, which describe how well they did it. Neither tells you whether you currently hold the capability to deliver what is on your roadmap.
The conventional answer is a self-assessment survey against a framework. These are cheap to run and produce data of very limited value, because self-rating measures confidence at least as much as competence, and it measures it inconsistently across individuals.
The alternative is structured interviews with an assessor, which produces far better evidence and does not scale.
How it works
Capable runs adaptive interviews using the STAR structure (situation, task, action, result), with the conversation driven by Claude and mapped to SFIA 9.
Because the interview adapts, it can follow up on a vague answer, probe for the specific contribution a person made within a team effort, and tell the difference between someone who led a piece of work and someone who was adjacent to it. That distinction is exactly what a survey cannot make and what a skilled human interviewer makes constantly.
Reports are evidence-based: each rating is tied back to what the person actually described. This matters for credibility. An assessment that produces a number is easy to dispute; one that produces a number alongside the evidence behind it can be discussed sensibly, including where the person disagrees.
Neurodiversity-first
Conventional interviews measure interview performance. For a substantial number of capable technologists, that is a poor proxy for capability, and the gap is wider for neurodivergent people, who may be excellent at the work and disadvantaged by the format used to assess it.
Capable was designed around that from the beginning: a conversational format without time pressure, follow-ups that clarify, and assessment of the experience someone describes instead of how they perform while watched.
Where it fits
Capability assessment is usually the missing input to an AI workforce strategy. Agencies plan uplift programmes without knowing their starting point, then measure success by attendance. Knowing what your people can actually do, before and after, is what turns a training budget into a capability plan.
Next step
We can show you Capable running against your own material.
A demonstration using your documents and your terminology tells you more in an hour than a scripted walkthrough does.