Case 05
A career framework that makes AI an evaluable competency
My documentation team had no explicit career path. I designed a four-level framework, now approved by HR, in which working well with AI is assessed like any other skill.
- 4
- levels, with 27, 27, 31 and 34 skills
- 18
- functions of the role, with at most 12 active per level
- 2
- AI competencies, one function and one hard skill
My decisions 05
- Profiles describe what each level is; they are not job postings.
- Progression is non-linear: not every skill rises one level per step.
- The required level is the standard, not the ceiling, so Expert stays as room to grow.
- At most 12 active functions per level, and N/A means not evaluated at that level, never zero.
- Soft skills are taken word for word from the corporate catalog of 35 competencies.
Assisted execution 02
- Claude generated the three workbooks from one shared data module and checked programmatically that the soft skills match the catalog word for word.
- Claude drafted the level profiles for my review.
Context
My documentation team had no explicit career path. There were no defined levels, no statement of what each level is expected to do and no way to measure growth. I also wanted AI to stop being an extra that some people use and others do not.
What I decided
- Profiles, not job postings. Each profile describes what a level is.
- Non-linear progression. Not every skill rises one level with each step.
- Required is not the ceiling. The required level is the standard to do the job. Expert stays as room to grow. The first draft put almost everything at Expert; I corrected it, and only two Manager functions sit at Expert: people leadership and strategy.
- At most 12 active functions per level. N/A means “not evaluated at this level”, not a score of zero.
- Two deliberate overlaps for the Manager. The Manager’s functions favor strategy and administration, with two tactical overlaps kept on purpose: documentation development and AI governance, both at Advanced.
- Soft skills verbatim. They come word for word from the corporate catalog of 35 competencies.
- Files by audience. What the team sees carries no noise meant for managers.
How it works
Four levels, each with its expected profile:
| Level | Functions | Hard skills | Soft skills | Total |
|---|---|---|---|---|
| Junior Technical Writer | 12 | 10 | 5 | 27 |
| Intermediate Technical Writer | 12 | 10 | 5 | 27 |
| Senior Technical Writer | 12 | 14 | 5 | 31 |
| Documentation Manager | 12 | 14 | 8 | 34 |
AI is part of the role. The function “AI Governance & Content Enablement” goes from Basic at Junior to Advanced at Senior and Manager. The hard skill “AI Tooling & Prompt Engineering for Documentation” is assessed like docs-as-code or API documentation standards.
Each skill has an expected level (Basic, Intermediate, Advanced or Expert) and behavior descriptions per level. Measurement uses the 360° portal’s scoring model, and the gap between the expected and the assessed level feeds each person’s development plan.
The design drew on public Technical Writing handbooks and industry career guides. It produced three workbooks from one shared data source: the framework, for the team; the mapping by role and seniority, with calculated transitions; and a self-contained Manager reference with a path for Seniors interested in management.
What it produced
A framework approved by HR, which HR is now implementing. Documentation is one of three areas with fully defined roles in the 360° portal: 4 of its 15 roles. Using AI well is now an evaluable competency of the Technical Writer role.
What it does not prove yet
Nobody has been evaluated with the framework at scale yet. I have no data on how many people it has assessed, on promotions or on the average gap between expected and assessed levels.
What I would do next
Run the first full evaluation round with the framework in the portal. Then track the average gap per skill, especially the two AI competencies, and see whether development plans close it.