Date · Wednesday, 6 May 2026, 12:00 to 13:00 CEST
Hosted by · UNV and UNICEF
Session page · UNOG learning
Speakers
- Rolf Reinhardt · Senior Account Director, LinkedIn · Bio
- Tom Frohner · Senior Customer Success Manager, LinkedIn · Bio
- Godwin Otim · People and Culture Specialist, UNICEF · Bio
AI for Your Career: Practical Tools and Prompts for Career Management is a hands-on session that explores how AI can support you throughout your career journey. Through LinkedIn Learning, delivered segments, you’ll see how AI can help you better understand your profile, tailor applications, and prepare for interviews, while keeping your voice authentic. A recruiter perspective adds helpful insight from the other side of the process. You’ll leave with simple, reusable prompts and a clear understanding of how to use AI effectively without over-relying on it.
Key takeaways
- Before applying, run the AI fit analysis to get a concrete gap map. not to decide whether you should apply, but to understand what is demonstrated, what is partial, and what is missing. Use it to calibrate whether to apply now or build a 30-day sprint first.
- The recruiter is not assessing your effort. They are assessing your evidence. Quantified impact, scale of responsibility, and consistency across materials are what survives scrutiny.
- AI roleplay enables deliberate practice before real interviews. Twenty rounds of simulated practice builds genuine confidence, sharper answers, real repetitions, mistakes made in private.
- Block development actions in your calendar using an AI-generated sprint plan in ICS format. Scheduling turns intention into execution.
- Optimize your LinkedIn profile in UN-aligned language. UN recruiters increasingly use AI hiring assistants, which are fed by your profile text.
- When prompting AI for learning, specify your proficiency level, ask for a structured roadmap, and push it deeper: “What would an expert know that a beginner wouldn’t?”
- Never send AI output without reading it. AI can be wrong, and you are responsible for what you send. ---
Sina Weinhuber
Sina set up the session as a working clinic rather than a lecture, asking participants to identify their single biggest challenge upfront, role fit uncertainty, unclear skill gaps, or difficulty sticking to a development plan, and carry that through as a lens for each segment. She introduced the 3 E’s of development framework: Experience (learning by doing), Exposure (learning through others), and Education (structured courses and AI tools). Her key contribution was pointing out the gap that most AI tools fill only the Education dimension, while real skill growth happens through a cycle of practicing in real contexts, getting feedback, and adjusting. She pushed participants to always follow a course or AI output with an immediate question: where and with whom will I apply this next week?
Rolf Reinhardt
Rolf demonstrated two practical AI prompts. The first runs a structured fit analysis against a job vacancy: paste your CV and the job description into any generative AI, and a structured prompt walks the AI through input analysis, fit scoring, gap identification, specific development actions, and translation into UN-aligned language for your LinkedIn profile. The purpose is not to get a green light to apply but to generate a concrete gap map, what’s clearly demonstrated, what’s partially aligned, what’s missing. He made the practical point that job descriptions in the UN system tend to be long and detailed, which is actually an advantage for AI analysis: more context means more specific output. He also noted that people socialized as male tend to apply even without full fit, while people socialized as female tend to hold back, the fit analysis helps calibrate that gap in both directions.
The second prompt converts that gap analysis into a 30-day professional development sprint. It generates a calendar-ready plan mixing learning tasks, mentoring check-ins, and coaching actions, all tailored to the specific vacancy. The output can be exported as an ICS file and imported directly into Outlook or Google Calendar, blocking the time before anyone else fills it. He also made a point about AI hygiene: always read the output before you send it. AI can be wrong, and you are responsible for what you send.
Tom Frohner
Tom showed two LinkedIn Learning features. The first is the AI Coach: given a skill and a stated proficiency level, it searches the LinkedIn Learning library and returns a structured roadmap with beginner-to-advanced course proposals. His three principles for effective prompting carry beyond LinkedIn Learning: be specific about your level, ask for a structured roadmap rather than a list of tips, and challenge the AI to go deeper by asking “what would an expert know that a beginner wouldn’t?” The second feature is AI roleplay, available to all LinkedIn users. You set up a scenario, a tough stakeholder conversation, a job interview, a performance discussion, define the AI persona’s characteristics, and then practice the conversation in real time, with no script and no multiple choice. The AI responds dynamically, pushes back when appropriate, and delivers a strength-and-improvement analysis at the end, with course suggestions tied to the development areas it identified. He cited deliberate practice research: 70% of skills are learned by doing, not watching, and AI roleplay enables that at scale, in private, with immediate feedback. The practical upshot: someone who has done 20 rounds of AI interview practice walks into the room with genuine confidence, not just theoretical preparation.
Godwin Otim
Godwin brought the recruiter-side perspective. He framed AI as a mirror: it sharpens and clarifies your self-presentation, but it cannot create substance that isn’t there. At the long-listing stage, recruiters are using AI to score applicants against minimum requirements and reduce a pool of 500 to a manageable shortlist, it processes volume, it does not understand fit. Every stage after that, shortlisting, interviewing, reference checks, involves increasing human judgment. The closer the process gets to a hiring decision, the less your AI-polished materials matter and the more your actual evidence of experience matters.
His core message: the recruiter is not going to look at your effort. They are going to look at your evidence. Signals that increase credibility include quantified achievements (budget managed, headcount led, percentage reductions delivered), clear career progression, and tight consistency between what you claim and what you demonstrate when probed. Signals that reduce credibility include generic AI-heavy language that sounds like everyone else’s application, inflated claims without specific proof, keyword stuffing without substance behind it, and inconsistencies in timelines or titles across the application. He closed with a diagnostic question worth holding: if a recruiter stripped away all the polished language from your CV today, what real evidence would still remain?
Frameworks and models
| Name | What it stands for | How to use it |
|---|---|---|
| STAR Method (referenced, no separate framework page) | Situation, Task, Action, Result | Structure examples of past experience in CVs, cover letters, and interviews to demonstrate specific evidence rather than generic claims. (Considered for promotion to a framework page but not promoted: STAR is closely related to the R-CAR and Skills-in-Use CV Pattern pages, which cover the same competency-statement structure with more operational detail. Use those.) |
| 3 E’s of Development | Experience, Exposure, Education | Plan professional development across all three dimensions: learning by doing (new tasks), learning through others (mentors, feedback), and structured learning (courses, AI tools). Most AI tools cover only the Education layer; plan for all three |
| Career Gap to Sprint Workflow | Two-prompt AI workflow: structured fit analysis, then 30-day calendar-blocked development sprint | Run prompt 1 with your CV and a target JD to produce a gap map; run prompt 2 to convert the map into an ICS-exportable sprint plan. Re-run prompt 1 on day 28 to test whether the gap closed |
| AI Roleplay for Skill Practice | Practice routine using AI personas to rehearse high-stakes conversations before they happen for real | Define scenario, persona, and rubric; run multiple rounds; collect structured feedback; iterate. Free for all LinkedIn users; same approach works on any general-purpose AI |
| AI Prompting for Learning | Three principles for getting useful learning support from any AI: be specific about your level, ask for structure, push it deeper | Use whenever you ask an AI to help you learn something. Pair the AI output with an immediate Experience commitment to keep it from staying in the Education-only layer |
| Evidence vs Polish Diagnostic | Recruiter-side question: if all polished language was stripped from this CV, what evidence would remain? | Run before submitting any application, after the Career Gap to Sprint Workflow and the Third Eye Principle review |
Resources
| Resource | What it is / What it’s for | Link |
|---|---|---|
| LinkedIn Learning AI Coach | Personalized course roadmap builder based on skill and proficiency level; part of LinkedIn Learning | Available via LinkedIn Learning (license required for Career Hub features) |
| LinkedIn Learning AI Roleplay | Scenario-based practice with dynamic AI personas; available to all LinkedIn users; English, German, French | Available in LinkedIn Learning |
| arena.ai | Runs the same prompt across multiple LLMs simultaneously; useful for comparing outputs | https://www.arena.ai |
| LinkedIn Premium career resources | Overview of LinkedIn Premium career tools | https://premium.linkedin.com/careers/career |
Last updated 2026-05-10.