Discovering Strengths Using AI - How 10 Strategic Questions Revealed My Squiggly Constellation

Sometimes career development feels like trying to navigate without a map. You think you’re heading in a straight line, only to find the path bends, loops, and squiggles its way in unexpected directions. That’s why I’ve always connected with Helen Tupper and Sarah Ellis’s Squiggly Careers approach. Careers aren’t linear, and that’s perfectly okay.

So when I spotted their latest Squiggly Careers Skills Sprint × AI, I jumped straight in. The timing couldn’t have been better: a week off between jobs, a fascination with AI, and a desire to keep learning. The stars aligned (quite literally), as I’d soon discover in the form of a strengths constellation.

Let's begin our strength discovery...

The first sprint focused on strengths: the things that give you energy, help you do your best, and shape what you’re known for. To help uncover these, the sprint suggested a 10-minute exercise using AI. This was the prompt Helen and Sarah shared, along with a choice of free generative AI tools (Claude, ChatGPT or Copilot):

Act as a squiggly career coach. Be high care and high challenge in your approach. Ask me 10 questions, one at a time, to help me explore what my strengths might be. At the end, use my answers to create a summary of what my top strengths could be.

The prompt turned Claude (my tool of choice for creative copy) into a “Squiggly Career Strengths Coach.” I thought it would be good to try a different tool as a regular user of both Copilot and ChatGPT. The idea was simple: Claude asked me 10 structured questions, one at a time, then used my answers to generate a personalised summary of my strengths. What struck me wasn’t just the end result but the quality of the questions themselves and how they built from one another.

The power of good questions

Claude didn’t just fire questions at me. It nudged me to go deeper, to share more detail, and to reflect on things I might otherwise skim over. The style felt surprisingly warm and personal, and at times almost therapeutic. Whether that’s down to the orange squiggle icon or the tone of its responses, it felt more like a conversation than filling out a form.

Here are a few of the questions that stood out (lightly adapted from my own session to keep them broadly useful to you):

• Flow moments: Think about a time when you were completely absorbed in what you were doing, so much so that you lost track of time. What exactly were you doing, and why did it engage you so much?

• Energy drainers: On the flip side, what types of work situations leave you feeling drained or frustrated? What specifically about them has that effect?

• Problem-solving style: When you face something complex with lots of moving parts, how do you naturally go about breaking it down?

• Learning curve: When you tackle something completely new, what helps you stick with it through the steep learning phase?

• Future identity: If in five years people came to you for one particular capability or expertise, what would you secretly hope it was?

  
  

Each question acted like a mirror, reflecting not just what I do but why I do it, and how it connects to the energy I gain or lose in my daily life and at work.

My constellation

At the end, Claude pulled my answers together into a strengths constellation. Not only did I get a neat HTML visualisation that mapped out my core strengths, but I was also able to use natural language to tailor the constellation to my own style and colour palette.

That aside, the constellation felt accurate but also aspirational, showing me not just who I am today but who I could lean into becoming.

Why AI works well here

There’s something powerful about using AI in this context. The speed, supportive tone, and accessible interface felt less judgemental than a human conversation, and the framing of Claude’s questions encouraged honesty and reflection.

Claude also replayed me back to myself. By presenting my own words in new ways, it pushed me to see patterns and insights I might have otherwise missed.

It also reminded me of some of the work I’ve been doing recently with Copilot Studio, where I’ve been building an agent that assesses an individual’s current skills and competencies, maps them against their roles, and predicts future skill gaps based on the impact of AI. The reliance on good data inputs is the same. Whether it’s my Copilot agent or Claude in this exercise, better questions lead to more valuable insights.

My takeaway from Day 1

Give it a go. Even if you’re sceptical, you’ll either learn something new about yourself or, at the very least, learn something about how AI can be used in more personal and reflective ways.

If you do try it, please feel free to share your constellation and see whether it shifts how you approach your work, your energy, and the skills you lean into.

This is just Day 1 of the sprint, and I hope to share more reflections as the week goes on. If you’d like to join in yourself, you can sign up via Amazing If and download the Skills Sprint × AI guide.

---

Credits

Author: Rob Anderton

Editorial: Rob Anderton / OpenAI. (2025). ChatGPT 5 [Large language model]. https://chat.openai.com/chat

Images: Claude (2025). Sonnet 4 [Large language model] https://claude.ai/ / Midjourney (2025). V7 [Image generation model] https://www.midjourney.com/ / OpenAI. (2025). ChatGPT 5 [Large language model]. https://chat.openai.com/chat

References

Inspiration and sign-up info: AmazingIf: https://www.amazingif.com/

Sprint Day 1 video: https://www.youtube.com/watch?v=XTD26WCqztA