Skip to content
K W
EU AI Act III(4)(b): High Risk Q4

Skills & Career Profile Agent

Build the skills inventory that every talent decision depends on.

Maintains skill profiles, matches them against role requirements, and shows possible career paths. EU AI Act high-risk classification applies.

Analyse your process
Airbus Volkswagen Shell Renault Evonik Vattenfall Philips KPMG

Skill extraction via AI, profile structure per rules, matching proposal

The agent extracts competencies from certificates and project experience via AI analysis, structures them deterministically into the company skill taxonomy and proposes role matches - the validation of competency level remains Human-in-the-Loop with employee and manager.

Outcome: According to BCG, 60 percent of HR departments assess their internal skill transparency as insufficient; for 87 percent of newly advertised positions, internal candidates would be qualified - but are not found.

13% Rules Engine
49% AI Agent
38% Human

The structural problem is not the capture but the gap between CV data and required role description:

The right candidate is sitting three floors below

The logistics team lead resigns. HR searches internally - and finds: no one knows who in the company has supply chain certifications, who brings project leadership experience, who has been waiting two years for exactly this chance. So the role is filled externally. Six months of search, six-figure costs, twelve months to full productivity. Three floors below sits a coordinator who meets all the requirements - but none of it is recorded anywhere.

This scenario is not an edge case. It is the normal state.

The invisible inventory

This agent follows the Decision Layer principle: each decision is either rule-based, AI-assisted, or explicitly assigned to a human.

93 percent of executives consider the shift from rigid job descriptions to skill-based models important or very important. But only 19 percent of organisations are ready to make that shift in practice (Deloitte, 2025). Only 16 percent use skills data for personnel decisions to any meaningful extent.

The result is a massive information gap. Organisations know which positions they have. They do not know which capabilities they have.

What HR systems typically know           What they do not
──────────────────────────────────────   ──────────────────────────────────
Job title                                Actual skills
Department                               Project experience outside the role
Last training (date)                     Informally acquired capabilities
Formal qualifications                    Willingness to develop
Pay grade                                Employee's career goal

This gap has direct economic consequences. According to LinkedIn Global Talent Trends 2024, internal hires are 18 to 20 percent cheaper than external hires, and employees from internal mobility programmes stay 41 percent longer with the company. Internal candidates reach full competence 20 percent faster than external new hires.

But mobility requires transparency. And transparency requires a living skills inventory.

Why skills profiles age before they are finished

Most companies have built skills profiles at some point. The problem is not the initial capture - it is the shelf life.

A typical skills profile emerges in the annual review conversation. The manager and the employee sit together, rate skills on a scale, document the result. Twelve months later, they repeat the ritual. In between, the employee has led a migration project, earned a cloud certification, and steered a team through a reorganisation. None of it ends up in the profile.

On top of that sits a systematic distortion. 48 percent of employees believe performance reviews are influenced by bias (SHRM). Self-assessments are notoriously unreliable - the Dunning-Kruger effect ensures that those who have the least competence most often overestimate it. Managers frequently rate based on the last impression rather than actual skill development over months.

A profile built on an annual subjective snapshot is not a skills inventory. It is an estimate with an expiry date.

From estimate to composite profile

The Skills and Career Profile Agent changes the underlying logic: instead of an annual assessment, a profile emerges that draws from multiple sources and updates continuously.

Source 1 - self-assessment. The employee rates their own skills. This stays important - not for accuracy, but because it promotes ownership and reflection. But it is one source among several, not the only one.

Source 2 - manager validation. The direct manager confirms, corrects, or supplements the assessment. This step stays with a human - it requires context no system provides.

Source 3 - documented evidence. Certifications, completed training, project participation, role changes. This data already exists in various systems - LMS, project management, HR master data. The agent consolidates it and maps it against the competency framework.

The composite profile is not perfect. But it is systematically better than any single source. And it ages more slowly, because new evidence flows in continuously rather than being captured once a year.

Career paths: from wishful thinking to gap analysis

A skills profile alone changes nothing. It becomes relevant only when matched against a target role.

Current profile         Job architecture          Result
┌────────────┐         ┌────────────┐         ┌────────────────┐
│ Skills     │         │ Roles with │         │ Reachable      │
│ Experience │────────▶│ require-   │────────▶│ positions      │
│ Certificates│         │ ments      │         │ + gap per role │
└────────────┘         └────────────┘         └────────────────┘
      A+H                    R                       A

A = Agent    H = Human    R = Rules

The agent calculates which internal roles are reachable based on the current profile - and what concretely is missing. Not as a vague recommendation, but as a measurable gap: three skills at level 3, one certification, eight months of leadership experience.

This is the difference between “you could think about project management” and “for the project lead role in unit X, you are missing: PRINCE2 certification, experience with budgets over EUR 500,000 (USD 545,000), and your stakeholder management is at level 2 instead of the required level 4.”

Employees who see clear career perspectives stay. 85 percent feel more motivated when they can see concrete development paths within the organisation. The question is whether these paths are documented or only promised.

Internal job matching: solving the passive talent problem

In most organisations, employees learn about internal openings via the notice board - physical or digital. Those who miss the posting or do not consider themselves qualified do not apply. The best internal candidates are often the ones who are not actively searching.

The agent reverses the logic. For every new internal posting, profiles are automatically matched. Suitable candidates receive a notification - not as an assignment but as information: this role matches your profile, here is the calculated match score, here are the remaining gaps.

Whether the employee applies remains their decision. Whether they get the role is decided by the manager and HR. The agent ensures the information flows - not that a decision is made.

The regulatory frame: high-risk for a reason

Skills profiles that influence internal placements fall under EU AI Act Annex III(4)(b) - task assignment based on personal traits. This is not a bureaucratic hurdle. It is an appropriate classification. Because when an algorithm co-determines who is considered for which role, it has direct consequences for careers.

The obligations are concrete: risk management system. Transparency about data sources and matching logic. Human oversight of placement decisions. Logging of every decision step. The right of affected persons to see and contest automated assessments.

Works council co-determination rights apply as soon as skills profiles serve as selection criteria for internal transfers. That means: works council agreement before the first profile match runs. Not afterwards.

Organisations that see these requirements as a brake miss the point. The documentation obligation forces exactly the transparency that makes the process fairer than any undocumented manager decision behind closed doors.

The infrastructure effect

A skills profile system does not operate in isolation. The matching engine that compares internal candidates against role requirements is the same mechanic used by the Succession Planning Agent for successor candidates. The gap analysis feeds the Learning Path Recommendation Agent. The career path calculation forms the foundation for the Promotion Process Agent. The profile enrichment logic - combining data from multiple sources into a consistent picture - becomes the reusable pattern for every agent that works with employee data.

Each additional agent that taps into this infrastructure makes it more accurate. More data sources, better validation, higher profile quality. The Skills and Career Profile Agent is not a single product. It is the foundation of the talent infrastructure.

Internal roles fill from a current skills picture, not from the manager’s Rolodex

Roles are filled internally because it becomes visible who has the skills - not only who the manager happens to know. Employees see concrete development paths rather than vague promises. HR leadership has a current picture of the skills landscape - not an annual review table that is outdated the day after it is created. And when the EU AI Act takes full effect from August 2026, the compliance architecture is already in place, because it was designed in from the start. (US: NYC Local Law 144 and similar emerging state-level rules on automated employment tools create comparable transparency obligations.)

Micro-Decision Table

Who decides in this agent?

8 decision steps, split by decider

13%(1/8)
Rules Engine
deterministic
49%(4/8)
AI Agent
model-based with confidence
38%(3/8)
Human
explicitly assigned
Human
Rules Engine
AI Agent
Each row is a decision. Expand to see the decision record and whether it can be challenged.
Collect skill self-assessment Request and record employee's self-reported skills and proficiency Human

Employee self-assessment is the starting point for profile building

Decision Record

Decider ID and role
Decision rationale
Timestamp and context

Challengeable: Yes - via manager, works council, or formal objection process.

Integrate certification data Import current certifications from certification tracking system AI Agent

Automated import from Certification Tracking Agent

Decision Record

Model version and confidence score
Input data and classification result
Decision rationale (explainability)
Audit trail with full traceability

Challengeable: Yes - fully documented, reviewable by humans, objection via formal process.

Import training completions Update skill indicators based on completed training AI Agent

Automated inference from training completion records

Decision Record

Model version and confidence score
Input data and classification result
Decision rationale (explainability)
Audit trail with full traceability

Challengeable: Yes - fully documented, reviewable by humans, objection via formal process.

Request manager validation Ask manager to confirm or adjust employee's skill assessment Human

Manager validation adds calibration to self-reported data

Decision Record

Decider ID and role
Decision rationale
Timestamp and context

Challengeable: Yes - via manager, works council, or formal objection process.

Map skills to competency framework Align reported skills with organisational taxonomy AI Agent

AI-assisted mapping with human review for ambiguous cases

Decision Record

Model version and confidence score
Input data and classification result
Decision rationale (explainability)
Audit trail with full traceability

Challengeable: Yes - fully documented, reviewable by humans, objection via formal process.

Track proficiency changes Update proficiency levels based on new evidence AI Agent

Automated tracking from assessments, certifications, and training

Decision Record

Model version and confidence score
Input data and classification result
Decision rationale (explainability)
Audit trail with full traceability

Challengeable: Yes - fully documented, reviewable by humans, objection via formal process.

Record career aspirations Capture employee's career goals and development preferences Human

Employee-driven input from development conversations

Decision Record

Decider ID and role
Decision rationale
Timestamp and context

Challengeable: Yes - via manager, works council, or formal objection process.

Provide profile access Make profile visible to employee with correction capability Rules Engine

Transparency requirement - employees must see and can contest their data

Decision Record

Rule ID and version number
Input data that triggered the rule
Calculation result and applied formula

Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.

Decision Record and Right to Challenge

Every decision this agent makes or prepares is documented in a complete decision record. Affected employees can review, understand, and challenge every individual decision.

Which rule in which version was applied?
What data was the decision based on?
Who (human, rules engine, or AI) decided - and why?
How can the affected person file an objection?
How the Decision Layer enforces this architecturally →

Does this agent fit your process?

We analyse your specific HR process and show how this agent fits into your system landscape. 30 minutes, no preparation needed.

Analyse your process

Governance Notes

EU AI Act III(4)(b): High Risk
Classified as high-risk under the EU AI Act, Annex III, Section 4(b) - the agent maintains personal data used for task assignment decisions based on personal traits. Conformity assessment mandatory. Employees must have full transparency into their skills profiles and the right to contest inaccurate data. Article 26(7) requires informing worker representatives. GDPR rights of access, rectification, and erasure apply directly. The agent records and structures - it does not evaluate or rate employees. The Decision Layer decomposes every process into individual decision steps and defines for each: Human, Rules Engine, or AI Agent. Every decision is documented in a complete decision record. Affected employees can understand and challenge any automated decision.

Assessment

Agent Readiness 51-58%
Governance Complexity 71-78%
Economic Impact 54-61%
Lighthouse Effect 64-71%
Implementation Complexity 51-58%
Transaction Volume Quarterly

Prerequisites

  • Organisational competency framework with skills taxonomy
  • Self-assessment and manager validation workflows
  • Integration with certification tracking, LMS, and performance systems
  • Employee-facing profile interface with correction capability
  • EU AI Act conformity assessment for high-risk classification
  • Works council agreement on skills profiling system
  • Data Protection Impact Assessment for personal attribute profiling
  • Defined data quality and freshness standards

Infrastructure Contribution

The Skills & Career Profile Agent is the data backbone for the entire talent management layer. Workforce Planning, Succession Planning, Training Needs Analysis, Learning Path Recommendation, and Promotion Process agents all depend on skills data. Building this infrastructure is a prerequisite for data-driven talent management. Builds Decision Logging and Audit Trail used by the Decision Layer for traceability and challengeability of every decision.

What this assessment contains: 9 slides for your leadership team

Personalised with your numbers. Generated in 2 minutes directly in your browser. No upload, no login.

  1. 1

    Title slide - Process name, decision points, automation potential

  2. 2

    Executive summary - FTE freed, cost per transaction before/after, break-even date, cost of waiting

  3. 3

    Current state - Transaction volume, error costs, growth scenario with FTE comparison

  4. 4

    Solution architecture - Human - rules engine - AI agent with specific decision points

  5. 5

    Governance - EU AI Act, works council, audit trail - with traffic light status

  6. 6

    Risk analysis - 5 risks with likelihood, impact and mitigation

  7. 7

    Roadmap - 3-phase plan with concrete calendar dates and Go/No-Go

  8. 8

    Business case - 3-scenario comparison (do nothing/hire/automate) plus 3×3 sensitivity matrix

  9. 9

    Discussion proposal - Concrete next steps with timeline and responsibilities

Includes: 3-scenario comparison

Do nothing vs. new hire vs. automation - with your salary level, your error rate and your growth plan. The one slide your CFO wants to see first.

Show calculation methodology

Hourly rate: Annual salary (your input) × 1.3 employer burden ÷ 1,720 annual work hours

Savings: Transactions × 12 × automation rate × minutes/transaction × hourly rate × economic factor

Quality ROI: Error reduction × transactions × 12 × EUR 260/error (APQC Open Standards Benchmarking)

FTE: Saved hours ÷ 1,720 annual work hours

Break-Even: Benchmark investment ÷ monthly combined savings (efficiency + quality)

New hire: Annual salary × 1.3 + EUR 12,000 recruiting per FTE

All data stays in your browser. Nothing is transmitted to any server.

Skills & Career Profile Agent

Initial assessment for your leadership team

A thorough initial assessment in 2 minutes - with your numbers, your risk profile and industry benchmarks. No vendor logo, no sales pitch.

30K120K
1%15%

All data stays in your browser. Nothing is transmitted.

Frequently Asked Questions

Does the agent assess or rate employees' skills?

No. Skills data comes from employee self-assessment and manager validation. The agent structures, maps, and maintains this data - it does not generate proficiency ratings independently.

Can employees see and correct their own profiles?

Yes. Transparency is both a design principle and a regulatory requirement. Employees have full access to their skills profiles and can flag inaccuracies for correction.

Why is this high-risk under the EU AI Act?

Skills profiles can be used for task assignment and career decisions based on personal traits - which the EU AI Act classifies as high-risk (Annex III, Section 4(b)). This does not prevent building the system - it defines the governance standard it must meet.

What Happens Next?

1

30 minutes

Initial call

We analyse your process and identify the optimal starting point.

2

1 week

Discover

Mapping your decision logic. Rule sets documented, Decision Layer designed.

3

3-4 weeks

Build

Production agent in your infrastructure. Governance, audit trail, cert-ready from day 1.

4

12-18 months

Self-sufficient

Full access to source code, prompts and rule versions. No vendor lock-in.

Implement This Agent?

We assess your process landscape and show how this agent fits into your infrastructure.