Blog
UX Testing with Claude Code — Meetup Recording
Recording from our April 21 PoDojo meetup: an end-to-end UX-testing workflow inside Claude Desktop, and what 'agentic' actually means in practice.
Yesterday we kicked off the PoDojo Studio Berlin meetup series with UX Testing with Claude Code — your personal UX research agent at work. Online, in Claude Desktop, in front of a live audience. Below is the unedited screen recording.
Why we ran this
User testing used to mean a doc, a recruiting spreadsheet, moderated sessions on your calendar, transcripts in one tool, synthesis in another. Weeks from the first idea to an insight you can act on. The agent doesn't just speed up the steps — it dissolves the seams between them.
The question we wanted the room to sit with: what does a UX-research loop look like when the agent is the runtime?
What's in the recording
- The Claude Code → Claude Agent reframe — why thinking in agents changes what you'll try, even before you touch any tooling.
- What Claude Code actually is — and how it's different from the Claude chat window most people have already used.
- A live Show & Tell on Claude Desktop — planning a test in a conversation, spinning up the test app, shipping it to a user, getting auto-transcribed feedback, iterating — without leaving the app.
- How auto-transcription + agent-driven analysis cut the slow, expensive parts of research.
- Where this fits beyond UX testing — across the whole product-development flow.
We're not building the next app you have to switch to — we're building our product so you can use it with your agent.
Hosted by Stefan Haas, Jochen Ade, and Hannah Carlotta Kelbel.
Next steps
- Join the next meetup — meetup.com/podojo-studio-berlin
- Rebuilding your own research loop? Drop us a line via the link at the bottom of this post — bring a real pain, we'll walk it through.
Unedited screen recording from the live session — expect the rhythm of a live call, not a produced video.