Agenda
Machine Learning Roundtable: ML in Game Making
Mathieu Nayrolles (Director of Applied AI/ML, Supercell)
Pass Type:
Academic Pass, Festival Pass, Game Changer Pass, Indie Pass, Start-Up Pass -
Get your pass now!
Track:
Programming, Production & Team Leadership
Audience Level:
All
Format:
Roundtable
Vault Recording: TBD
Audience Level: All
This is the table about making the game, not playing it. Coding agents, asset pipelines, automated QA, and everything in between. The question is no longer whether the tools work but where you let them touch your codebase and your art.
Agentic coding stopped being a novelty. What stuck, what got rolled back, what ended up on internal branches with a human gate and a checkpoint. Asset generation that blocks out a level but is still not final art. Review queues that grow faster than the code. Store disclosure rules that now care about what the player sees, not which assistant wrote the shader.
The goal is lived experience from the trenches. Not predictions. Not a vendor roadmap. Just what it feels like to make games with agents in the loop.
Agentic coding stopped being a novelty. What stuck, what got rolled back, what ended up on internal branches with a human gate and a checkpoint. Asset generation that blocks out a level but is still not final art. Review queues that grow faster than the code. Store disclosure rules that now care about what the player sees, not which assistant wrote the shader.
The goal is lived experience from the trenches. Not predictions. Not a vendor roadmap. Just what it feels like to make games with agents in the loop.
Takeaway
Attendees gain a grounded understanding of how ML and generative tools are used in game production today. This includes coding agents, asset pipelines, automated QA, data governance, and the review overhead that comes with all of it.
Rather than abstract potential, this session focuses on concrete wins, real costs, and what got rolled back. Attendees should leave with a better sense of what is worth trying next, what needs a human gate, and what is still not ready.
Rather than abstract potential, this session focuses on concrete wins, real costs, and what got rolled back. Attendees should leave with a better sense of what is worth trying next, what needs a human gate, and what is still not ready.
Intended Audience
This round table is aimed at ML practitioners in game development. This includes AI and ML engineers, technical directors, leads, tools engineers, and anyone shaping how games get built.
You do not need to be an ML expert, but you should care about how agents change the daily work. If you adopted, evaluated, or resisted these tools in the past year, you will have something to contribute.
You do not need to be an ML expert, but you should care about how agents change the daily work. If you adopted, evaluated, or resisted these tools in the past year, you will have something to contribute.