The role of the technical writer is changing as AI tools are used more and more. Organisations need people who can use them, evaluate them critically, build workflows around them, and guide teams in using them responsibly.
A career matrix is one of the most practical ways to define and develop those skills.
Below, we’ve outlined how to update a career matrix for Technical Writers so it addresses AI competencies. For full disclosure, our researching included using AI tools.
The matrix is structured around four common career levels:
- L1 – Associate / Junior Technical Writer
- L2 – Technical Writer / Mid-Level
- L3 – Senior Technical Writer
- L4 – Principal Technical Writer / Documentation Manager
And nine core competency areas:
- AI tool proficiency and workflow automation
- Prompt engineering and prompt systems
- AI output review and factual validation
- Creating chatbot-ready content
- Docs-as-code and engineering workflow
- Model evaluation and measurable quality
- Responsible AI, ethics, legal, and risk
- Cross-functional AI communication
- Leadership and enablement
Competency tables
Each table below defines the expected behaviours and outputs at each career level.
AI tool proficiency and workflow automation
At every level, writers should be able to use AI tools effectively. The definition of “effective” will change and grow as seniority increases.
| Level | Expectations |
| L1 | Uses approved AI tools for drafting, summarisation, and outline generation. Validates outputs before publishing. |
| L2 | Builds repeatable workflows and reusable prompt libraries. Improves cycle time without quality loss. |
| L3 | Designs multi-step AI pipelines. Integrates tools with document workflow and CI/CD. |
| L4 | Defines organisation-wide tooling standards, vendor strategy, adoption roadmap, and workflow governance. |
Prompt engineering and prompt systems
This section distinguishes between writing a good prompt (L1) and building and governing an enterprise prompt system (L4).
| Level | Expectations |
| L1 | Writes clear prompts with role, audience, output format, and constraints specified. |
| L2 | Uses few-shot prompting, reusable templates, validation loops, and prompt versioning. |
| L3 | Builds prompt frameworks for teams and measures prompt effectiveness using eval sets. |
| L4 | Defines enterprise prompt standards, governance, and reusable libraries across teams. |
AI output review and factual validation
AI-generated content introduces specific failure modes that require a deliberate review practice at every level.
| Level | Expectations |
| L1 | Detects hallucinations, tone drift, missing context, and factual errors against source material. |
| L2 | Builds editing checklists and verification workflows validated against source truth. |
| L3 | Creates team-wide QA frameworks and automated quality checks. |
| L4 | Defines organisation-level editorial standards and red-team review protocols. |
Creating chatbot-ready content
The way content is structured matters more than ever, as it affects RAG systems, knowledge bases, and other AI assistants.
| Level | Expectations |
| L1 | Writes modular content with consistent headings and metadata for retrieval. |
| L2 | Designs content for retrieval systems using semantic chunking and structured metadata. |
| L3 | Owns taxonomy, metadata schema, and docs-as-data workflows. |
| L4 | Defines enterprise information architecture standards for human and AI consumption. |
Docs-as-code and engineering workflows
This competency tracks the degree to which writers can operate fluently in engineering environments.
| Level | Expectations |
| L1 | Uses Markdown, Git basics, and PR workflows. |
| L2 | Owns CI/CD docs workflows, linting, and automated link checks. |
| L3 | Defines documentation infrastructure and AI integration pipelines. |
| L4 | Drives tooling strategy across product and engineering organisations. |
Model evaluation and measurable quality
AI-assisted documentation requires new approaches to quality assurance.
| Level | Expectations |
| L1 | Uses checklists and peer review rubrics to assess quality. |
| L2 | Builds evaluation datasets and test prompts; tracks pass rates. |
| L3 | Establishes regression tests and release gates for AI-assisted docs. |
| L4 | Owns evaluation scorecards and organisation-wide quality thresholds. |
Responsible AI, ethics, legal, and risk
This competency covers everything from following disclosure guidelines at the junior level to owning executive risk reporting at the principal level.
| Level | Expectations |
| L1 | Follows policy for disclosure, privacy, and safe use of AI tools. |
| L2 | Reviews outputs for bias, IP risk, and unsafe or misleading claims. |
| L3 | Develops responsible AI guidelines and disclosure frameworks for the team. |
| L4 | Owns ethics standards, compliance readiness, and executive risk reporting. |
Cross-functional AI communication
As AI features become more common in products, the ability to communicate model behaviour, limitations, and UX implications becomes a high-value skill.
| Level | Expectations |
| L1 | Collaborates with engineering and PMs to document AI features accurately. |
| L2 | Translates model behaviour and limitations into clear, user-facing documentation. |
| L3 | Advises product teams on AI UX, feature messaging, and model documentation. |
| L4 | Shapes executive AI narrative and cross-functional documentation strategy. |
Leadership and enablement
This competency tracks a writer’s ability to build capability in others.
| Level | Expectations |
| L1 | Shares learnings with peers; actively participates in team knowledge exchange. |
| L2 | Onboards writers into AI workflows and tools; documents best practices. |
| L3 | Builds team training curriculum and competency assessments. |
| L4 | Leads organisation-wide AI enablement strategy and capability roadmap. |
Roles and responsibilities by level
The table below summarises the primary focus of each career level across the competency areas listed above.
| Level | Primary focus |
| L1 | Uses AI tools safely and effectively. Validates outputs. Learns prompt fundamentals. Follows established guidelines. |
| L2 | Builds repeatable workflows. Manages prompt libraries. Improves quality controls. Mentors peers informally. |
| L3 | Designs systems and frameworks. Creates governance structures. Leads tooling decisions. Develops team capability. |
| L4 | Sets organisational standards. Drives AI strategy. Owns policy and risk. Reports to executive stakeholders. |
Promotion criteria
The following criteria define what a writer must consistently demonstrate to be considered for promotion to the next level. These are intended as a guide for managers and as a development roadmap for writers.
| Promotion to | Must consistently demonstrate |
| L1 (Entry) | Ability to use AI tools with appropriate validation. Follows team style and disclosure guidelines. Produces accurate first drafts with minimal revision. |
| L2 | Consistent AI-assisted drafting quality. Strong verification habits. Prompt reuse and modular content structure. Measurable reduction in revision cycles. |
| L3 | Ownership of systems and workflows. Development of prompt frameworks and retrieval-ready content architecture. Measurable quality improvement. Active mentorship of junior writers. |
| L4 | Organisation-wide standards and strategy ownership. Governance and compliance leadership. Executive influence. Demonstrable business impact tied to documentation quality. |
Role expectation mapping across competencies
Legend: N = Novice, W = Working, P = Proficient, E = Expert.
This mapping treats AI fluency as T-shaped: breadth expectations rise with seniority, but “Expert” depth is typically reserved for leads/principals and managers.
This table does not include all of the competencies suggested earlier.
| Competency | L1 | L2 | L3 | L4 |
| Model literacy for writers | W | W | P | P |
| Prompt engineering for documentation | W | P | P | P |
| Prompt chaining and agentic workflows | N | W | P | P |
| Data literacy and documentation of data | N | W | P | P |
| Model evaluation and quality metrics | N | W | P | P |
| Tool integration and structured outputs | N | W | P | W |
| API use and automation basics | N | W | P | W |
| LLM safety and security | N | W | P | P |
| Ethics, law, and responsible use | W | P | P | E |
| Content provenance and disclosure | W | P | P | E |
| AI documentation and governance | W | P | E | E |
Want to develop your team’s AI fluency?
Cherryleaf offers training in how to use AI in technical writing, as well as support recruiting AI-competent technical writers. We can also provide a spreadsheet version of this career matrix with the competency levels mapped out. Contact us to find out more.
Leave a Reply