The AI
Conference for Experimentation Leaders
the humans
in the loopMAY 11 & 12. 2026 LONDON UK
Join 200+ in-house leaders & innovators to explore and shape the future of how we build, ship, test, and work.
Extreme networking with workshops, training, talks, networking, parties, and ‘activation’ events
100+ table sessions matching you with the topics and people you care about.
Core audience: matching you with the topics and people you care about.
AGENDA SUMMARY
Location
- Day 1 workshops and preparty near Kings Cross area
- Day 2 conference at Big Penny Social, 5 stops (15min) up Victoria Line from Kings Cross
SUNDAY
Pre-pre party
Casual get together
MONDAY
AI workshop day
8:30am - 4/5pm with two primary workshops
VIP moderator/speaker cocktail party
Official Party
open to all
TUESDAY
Conference day
9am→11pm
- keynote talks
- networking table sessions
- lightning-talks
- AI poster session
- party
WEDNESDAY
Unofficial activity day
(optional) with special events/tours/fun
Full agenda coming soon
We've designed the entire agenda to engage and connect you with peers:
- 80% in-house/brand-side attendance
- Workshops, Dinners, and Masterclass events based on core roles
- Thoughtful parties and activation events
- Talks by leading in-house practitioners
- 100+ round tables, matching you with the topic and people you care about
The Core Topics We’re Scoping:
Ring 1
The AI-Powered Growth Engine
For the Practitioners & Their Leaders:
AI FOR EXPERIMENTATION
Go beyond "using AI" and learn to build automated growth and experimentation workflows for efficiency gain. We'll showcase the real-world tools and tactics for applying AI across the entire lifecycle - from generating sharper insights at scale to automating analysis and accelerating your dev-test-learn loop.
Ring 2
Taming The
Black Box
For the Builders & Technical Strategists:
EXPERIMENTING ON AI
This is the frontier. How do you run rigorous experiments when the product is the AI? We're tackling the hard problems: testing non-deterministic systems, measuring the impact of LLM-powered features, and finding the right metrics when testing recommendation engines and other "black box" models.
Ring 3
The Ring-
master’s
View
For the Leaders and Change-Makers: EXPERIMENTING WITH AI
This is about strategy, governance, and ROI. Get pragmatic, business-focused frameworks for the big questions: How do you structure teams for AI innovation? How do you handle the ethics and reliability to protect your brand? And crucially, how do you measure the return on your AI investments?
Workshops
Day 1 -optional
A day before the conference we host a special Workshop/Masterclass Day. We have 2 workshops to participate in, one set in the morning, and the second set in the afternoon.
Building Customer Insights Systems with Claude Code
What you'll walk away with:
- A multi-agent analysis system that handles mixed-methods data (transcripts, survey verbatims, quantitative data)
- Agents with built-in verification and feedback loops - they check their own work against source data
- A system that cross-references findings across data sources and flags contradictions
- party
What you'll build:
- Parallelized analysis agents that work simultaneously
- Sequenced agents that synthesize only after findings are verified
- Citation and evidence requirements baked into every output
- Markdown files that give you consistent, repeatable results from your workflows every time
Plus:
Use Caitlin's data and agent files to start, or use your data if you're working on a secure computer.
Prerequisites:
- Claude Code installed and working in your terminal (setup instructions sent before workshop)
- Claude subscription (Pro/Max) OR Anthropic API account with credits
- Laptop with terminal access
- Ideal: a few different data types for practice (3+ customer call transcripts, <100 survey responses or short customer feedback entries)
by Caitlin
Sullivan
Experimenting with AI: From Vibes to Workflows
What you'll build:
- A structured breakdown of a real project. Outputs, sub-tasks, and explicit AI intervention points along with where to use tools
- A workflow map that puts reliable, consistent tools first, with AI filling the gaps between steps
- A set of prompt patterns you can use to produce more consistent and reliable outputs
Start with your own project, or use the sample workflow provided.
What you'll walk away with:
- A working mental model of how AI actually works (pattern matching, not reasoning), so you know exactly when to trust it, when to verify it, and when to skip it entirely
- A task decomposition method: break any project into steps small enough for AI to handle reliably and outputs clear enough for you to check
- Interaction techniques drawn from the same approaches inside leading tools like Claude Code, but under your control, with transparency of every step
- A way to decide where AI belongs in a workflow, and when it does not
- A leaner, more sustainable way of using AI. That means smaller models, fewer tokens, no unnecessary spend, better output
- Reusable workflow patterns you can apply to your own projects the same week
Prerequisites:
- No technical or coding experience required
- Bring a real project or work process you want to improve. For example, a content workflow, a research task, a recurring deliverable, anything you do regularly
- A laptop
- Optional: any existing docs, briefs, or templates related to that project (or use the samples provided)
by Iqbal
Ali
Keynote Speakers
.jpeg)
More to be added soon...
Moderators
.jpeg)
.jpeg)
The Circus Difference: Why This is a True Shake-Up.
We're bringing the Experiment Live (Austin) format to Europe, and expanding it. We’ve built the event we've always wanted to attend. That means a complete rethink of the conference format.
Ready to Join the SHOW?
Individual tickets
Team ticket (3+ tickets)
