Keynote
  1. 09:00–09:05
    Welcome — Day 2
    John Allsopp
  2. 09:10–09:40
    Keynote
    Jeremy Howard
  3. 09:40–09:58
    Craft in the Time of Agents
    Annie Vella
  4. 10:00–10:20
    What If You Never Needed an API Key Again? Building a Mesh LLM From Spare Compute
    Mic Neale
  5. 10:20–10:38
    Towards Long-Horizon Tasks
    Zixuan Li
  6. 10:50–10:55
    Morning keynote outro
    John Allsopp
AI Engineering
  1. 12:00–12:18
    Deploying AI at the Edge: Model Compression and Hardware-Aware Optimization
    Shivay Lamba
  2. 12:20–12:38
    When a Small Language Model Beat Our LLM in Production
    Avni Bhatt
  3. 12:40–12:58
    Multi-Model Collaboration with Claude Code: How to Measure What Actually Works
    Jack Rudenko
  4. 13:00–13:18
    Edge AI with Direct Device Control
    Jeremy Kelaher
  5. 13:20–13:38
    COBOL and AI: Building a Self-Serve Knowledge Layer for 2,000 Batch Jobs
    Matthew Gillard
  6. 13:40–13:58
    Legacy Software + Agentic Discovery
    Chris Rickard
  7. 15:00–15:18
    Why LLMs Fall for Stories (And 5 Production Patterns That Actually Stop Them)
    Mal Curtis
  8. 15:20–15:38
    Hacking the Model: AI Red Teaming in Practice
    Pas Apicella
  9. 15:40–15:58
    Why Most AI De-Identification Fails in Production, And How We Built One Lawyers Actually Trust
    Moin Zaman
  10. 16:00–16:18
    Are Your AI Agents Secure? Defending the Privileged Agent
    Daizen Ikehara
  11. 16:20–16:38
    Your Agents Pass Every Benchmark—Then Memory Breaks Them in Production
    Ananya Roy
  12. 16:40–16:58
    AI Agents Are Distributed Systems
    Lovee Jain
Software Engineering
  1. 12:00–12:18
    Spec driven AI development - A Real World Perspective
    Nick Beaugeard
  2. 12:20–12:38
    Building SDKs in the Agentic Era
    Mark McDonald
  3. 12:40–12:58
    AGENTS.md is the wrong conversation
    Jakub Riedl
  4. 13:00–13:18
    Engineering without reading code
    Ben Taylor
  5. 13:20–13:38
    The Death of Documentation
    Josh Gillies
  6. 13:40–13:58
    Designing Inference-Native Systems
    Sajjad Kamal
  7. 15:00–15:18
    Agentic SAST: Building an AI Pipeline for Rule Synthesis and Root-Cause Vulnerability Analysis
    Danila Sashchenko
  8. 15:20–15:38
    AI After an Apocalypse
    Simon Knox
  9. 15:40–15:58
    Stop vibing your agents to production: applying ML discipline to agent development
    Justin Barias
  10. 16:00–16:18
    Fully Automated Luxury Gay Space Engineering
    Daniel Rodgers-Pryor
  11. 16:20–16:38
    12TB of AI coding agent logs - what works, what fails
    Dave Slutzkin
  12. 16:40–16:58
    Slop is a standards problem
    David Lewis
Leadership
  1. 12:00–12:25
    Not Everything Needs an LLM
    Dave Hall
  2. 12:30–12:55
    The AI Tax and "legal" ways to minimise it
    Krishna kanth Mundada
  3. 13:00–13:25
    Your engineers aren't afraid of AI. They're afraid of becoming junior again.
    Andy Kelk
  4. 13:30–13:55
    Panel: Engineering Reality
    AJ Fisher, Dave Hall, Krishna kanth Mundada, Andy Kelk
  5. 15:00–15:25
    From AI Survey to Production: What the Readiness Gap Actually Looks Like
    Dr Christian Dandre
  6. 15:30–15:55
    What We Learned Taking a Culture-First Approach to AI Adoption at scale
    Eric Grigson, Paul Hughes
  7. 16:00–16:25
    Beat Burnout, Find Flourishing: The AI Edition
    Navin Keswani
  8. 16:30–16:55
    Panel: Culture & People
    Andrew Murphy, Dr Christian Dandre, Eric Grigson, Paul Hughes, Navin Keswani
Hallway
  1. 11:15–11:35
    Build Agent-Powered Workflows on Notion Developer Platform
    Esti Frischling
  2. 11:40–12:00
    Why Engineering Velocity Still Breaks
    Aldo Schumann
  3. 14:10–14:30
    Stop Rebuilding Authorisation: Accelerating AI Agent Development
    Daizen Ikehara, Kevin Akermanis
  4. 14:35–14:55
    Hacking the Model: AI Red Teaming in Practice
Workshop
  1. 12:00–14:00
    Observability and Evaluation for LLM Apps and Agentic AI with Langfuse
    Muhammad Ali
All
  1. 07:30–08:45
    AI Leaders Roundtable Breakfast