Capstone project objective

You have completed 9 modules covering the entire employee lifecycle with AI: recruitment, selection, onboarding, engagement, training, evaluation, analytics, compliance and employer branding. Now it is time to apply it.

The capstone project consists of implementing AI in 3 HR processes of your organization (or a practical case if you do not have an organization available). This is not a theoretical exercise. It is about choosing 3 real processes, designing the AI solution, implementing it (at least as a pilot), and measuring the results.

Project criteria

• 3 processes: one from recruitment/selection (HR01-HR02), one from talent management (HR03-HR06), and one from analytics/compliance/branding (HR07-HR09).
• Real or simulated implementation: if you have access to an organization, real implementation with real data (anonymized when necessary). If not, simulation with fictional but realistic data.
• Measurement: each process must have at least 2 measurable success metrics before and after.
• Compliance: each process must include a legal risk analysis (GDPR, AI Act) and mitigation measures.
• Documentation: final deliverable in executive report format (not an academic essay, a document you can present to leadership).

Phase 1: HR-AI maturity assessment

Before deciding what to transform, you need to know where you stand. The maturity assessment gives you a snapshot of your starting point.

Assessment dimensions

1. Processes: which of your HR processes are manual, which are partially automated, and which already use AI?
2. Data: what HR data do you have? In which systems? What quality? Are they integrated?
3. People: what level of AI competency does your HR team have? Is there resistance to change?
4. Technology: what tools do you use? Do they have AI capabilities? Do they allow integrations?
5. Compliance: have you done DPIAs? Have you informed the works council? Do you have an AI policy?

Prompt for maturity assessment

"I am evaluating my HR department's maturity for implementing AI. I will give you information about each dimension and I need a score from 1 (basic) to 5 (advanced) with justification. Current processes: [description]. Available data: [description]. Team competencies: [description]. Tools: [list]. Compliance status: [description]. Generate: 1) Maturity scorecard (5 dimensions, score 1-5). 2) Overall score. 3) Top 3 strengths. 4) 3 most critical gaps. 5) Recommendation on where to start."

Score interpretation

• Score 5-10 (basic): start with the foundations. Clean data, train the team on basic AI, implement a single process with a general LLM (Claude or ChatGPT).
• Score 11-17 (intermediate): you already have a base. You can implement 2 or 3 processes with specialized tools. Focus on metrics and compliance.
• Score 18-25 (advanced): you are ready for predictive analytics, complex automations, and a comprehensive AI strategy for HR.

Phase 2: Selecting processes to transform

Not all processes are equal. Some generate more impact with less effort. Use a prioritization matrix to choose.

Impact/effort matrix

For each candidate process, evaluate:

• Impact (1-5): how much time/cost savings does it generate? How much does it improve employee/candidate experience? How much does it reduce risk?
• Effort (1-5): how much does it cost to implement (money, time, organizational change)? What level of data and technology does it require?

Candidate processes by category

Recruitment and selection (choose 1)

• AI job description writing (medium impact, low effort)
• AI-assisted CV screening (high impact, medium effort)
• Blind screening with automatic anonymization (high impact, medium effort)
• Structured interview guides with AI (medium impact, low effort)
• Personalized sourcing outreach (high impact, medium effort)

Talent management (choose 1)

• Automated onboarding: welcome kit + 30-60-90 plan (high impact, medium effort)
• Pulse surveys with sentiment analysis (high impact, low effort)
• Personalized learning paths (high impact, medium effort)
• Performance evaluation writing (medium impact, low effort)
• 360 feedback analyzed with AI (high impact, medium effort)

Analytics, compliance or branding (choose 1)

• Basic people analytics dashboard (high impact, medium effort)
• Salary equity analysis (high impact, medium effort)
• DPIA for AI system in HR (high impact, low effort)
• LinkedIn employer branding calendar (medium impact, low effort)
• Glassdoor review analysis (medium impact, low effort)

First-time recommendation

If this is your first AI project in HR, choose:

1. Job description writing + interview guides (quick impact, low risk)
2. Pulse surveys with sentiment analysis (high impact, low effort)
3. Employer branding calendar (visible, low effort, low risk)

Phase 3: Design and implementation

For each selected process, follow this structure:

1. Problem definition

• What specific problem does AI solve in this process?
• What is the current situation (baseline)?
• What is the expected result (target)?

2. Solution design

• What AI tool will you use (general LLM, specialized tool, automated workflow)?
• What data do you need and where is it?
• What is the step-by-step workflow?
• Where does the human intervene?

3. Compliance analysis

• Does this process involve automated decisions (GDPR Art. 22)?
• Is it classified as high-risk (AI Act Annex III.4)?
• Do you need a DPIA?
• Have you informed employees/representatives?

4. Pilot implementation

• Start with a small group (1 team, 1 department, 10 candidates).
• Document everything: what works, what does not, what adjustments you make.
• Collect user feedback (recruiters, managers, candidates, employees).
• Pilot duration: 2-4 weeks.

5. Adjustment and scaling

• Based on pilot results, adjust the process.
• If it works, scale gradually to the rest of the organization.
• If it does not work, document why and what you learned.

Phase 4: Results measurement

Each process must have clear success metrics measured before (baseline) and after (result) implementation.

Metrics by process

Recruitment/selection

• Job description writing time (before vs after)
• Application quality (rate of candidates passing first filter)
• Time-to-hire (days from opening to hire)
• Pipeline diversity (representation of underrepresented groups)
• Candidate experience score (CNPS)

Talent management

• Onboarding preparation time (HR hours per employee)
• New employee time-to-productivity
• Pulse survey response rate
• Engagement score (monthly evolution)
• Evaluation writing time (hours per manager)

Analytics/compliance/branding

• HR report generation time
• Compliance documentation completeness (% of checklist)
• Employer branding content reach and engagement
• Glassdoor rating (evolution)
• Organic applications (before vs after)

ROI formula

For each process, calculate:

ROI = (Savings generated + Impact value) / Solution cost

• Savings generated: hours saved x hourly cost of the employee who was doing them.
• Impact value: turnover reduction x replacement cost, time-to-hire reduction x cost per vacancy day, etc.
• Solution cost: tool subscription + implementation time + training.

Project template

Use this structure to document your project:

Executive report (2-3 pages per process, 8-10 pages total)

1. Executive summary (1 page): what we did, why, and what result we obtained.
2. Maturity assessment (1/2 page): scorecard with 5 dimensions.
3. Process 1: [name] (2 pages): problem, solution, compliance, pilot, metrics, result, learnings.
4. Process 2: [name] (2 pages): same structure.
5. Process 3: [name] (2 pages): same structure.
6. 12-month roadmap (1 page): next steps to scale the transformation.
7. Appendices: prompts used, data (anonymized), user feedback, compliance checklist.

3 examples of HR transformation with AI

Example 1: 80-employee SMB (professional services)

Processes chosen:

• Job description writing with Claude (before: 45 min/post, after: 10 min/post).
• Weekly 3-question pulse survey with automatic sentiment analysis (engagement rose from 6.2 to 7.1 in 2 months thanks to data-driven actions).
• LinkedIn calendar with 3 employer branding posts/week (organic applications rose 35% in 6 weeks).

Cost: 20 EUR/month (Claude Pro subscription) + 8 hours setup.

ROI: ~15 hours/month saved in HR tasks. ROI > 500% in the first quarter.

Example 2: 300-employee mid-size company (technology)

Processes chosen:

• AI CV screening + blind screening (pipeline diversity up 18%, time-to-hire down from 38 to 28 days).
• Automated onboarding with AI-generated welcome kit and 30-60-90 plan (time-to-productivity down from 5 to 3.5 months, 90-day retention up from 82% to 91%).
• People analytics dashboard with 8 key metrics + AI-generated monthly executive report (the CEO started making data-based talent decisions for the first time).

Cost: 200 EUR/month (tools) + 40 hours setup.

ROI: estimated savings of 180,000 EUR/year in avoided turnover. ROI > 3,000%.

Example 3: 2,000-employee multinational (regulated sector)

Processes chosen:

• AI-assisted performance evaluation (writing + bias detection). Evaluation preparation time down 60%. Systematic gender bias detected in review language and corrected.
• Turnover prediction with basic scoring model (23 high-risk employees identified, 18 intervened, 15 retained).
• Complete DPIA for 3 AI systems in HR + works council communication. GDPR and AI Act compliance verified before the August 2026 deadline.

Cost: 2,000 EUR/month (platform + legal advisory) + 120 hours setup.

ROI: 15 retentions x average replacement cost of 50,000 EUR = 750,000 EUR of protected value.

12-month HR transformation roadmap with AI

Months 1-3: Foundations

• Maturity assessment completed.
• HR team trained in basic AI use (level 2 from HR05 framework).
• 3 pilot processes implemented and measured.
• AI policy for HR drafted.
• DPIA for AI processes completed.

Months 4-6: Scaling

• Successful pilot processes scaled to the entire organization.
• People analytics dashboard operational with 5+ metrics.
• Monthly pulse surveys with automated analysis.
• Works council informed and consulted.
• Employer branding active on LinkedIn (3 posts/week).

Months 7-9: Deepening

• Predictive analytics: operational turnover prediction model.
• Training and development with AI-personalized paths.
• Continuous performance evaluation (weekly check-ins + quarterly evaluation).
• Fairness testing in all selection processes.
• Automated candidate nurturing for talent pool.

Months 10-12: Optimization

• ROI review of all initiatives. Scale what works, pivot what does not.
• System integration (ATS, HRIS, LMS, engagement) with AI layer.
• Annual people analytics report for leadership.
• Plan for year 2: what comes next?

Errores comunes in transformation projects

1. Starting with technology. "Let's buy Workday AI." No. Start with the problem. Which process hurts most? Which metric do you want to improve? The tool comes after.
2. Not involving the HR team. If recruiters or business partners do not understand AI, they will not use it. Or they will use it poorly. Training is not optional.
3. Measuring only efficiency. "We saved 20 hours per month." Good. But has hiring quality improved? Has engagement gone up? Efficiency without quality is dangerous.
4. No executive sponsor. A transformation project without CEO or CHRO support will die at the first budget obstacle.
5. Doing everything at once. 3 processes in 3 months is ambitious but manageable. 10 processes in 1 month is guaranteed disaster. Prioritize, pilot, scale.
6. Forgetting compliance. If you implement AI in selection without a DPIA or informing workers, the time savings you generate will be lost in sanctions and litigation.
7. Not celebrating quick wins. The first report AI generates in 5 minutes instead of 5 hours, the first candidate hired via automated nurturing, the first Glassdoor review responded to with empathy. Celebrating it generates momentum.

Project submission

Specialization closing

You have completed the HR and Talent with AI specialization at IAcademy. In 10 modules you have learned to apply artificial intelligence at every stage of the employee lifecycle:

• AI-assisted recruitment (HR01)
• Bias elimination in selection (HR02)
• Automated onboarding (HR03)
• Engagement and workplace climate (HR04)
• Training and development (HR05)
• Performance evaluation (HR06)
• People analytics (HR07)
• Employment compliance and AI Act (HR08)
• Employer branding (HR09)
• Transformation project (HR10)

AI does not replace the HR professional. It empowers them. It eliminates low-value tasks (processing paperwork, copying data, writing generic text) so you can dedicate your time to what really matters: people.

The talent market is more competitive than ever. HR professionals who master AI will have a decisive advantage. Not because AI is magical. Because AI will allow them to make better decisions, faster, with less bias, and with more data. And that translates into better hires, higher retention, more engaged employees, and a stronger organization.