Share the system you actually shipped — what broke, what you rebuilt, and what you would do differently. Every track is led by practitioners, for practitioners.
Apply to speak
We look for talks that go beyond surface-level overviews. Show us your architectural decisions, tradeoffs and reasonings behind your approach.
Theory is easy; shipping is hard. We want to hear what you learned building systems that actually run at scale.
If you've solved an infra or tooling problem in a novel or unexpected way, this is the audience that will appreciate it.
We look for speakers who have done the work themselves. Show us your hands-on experience, and why you’re the right person to explain this.
A decade of engineers, architects and researchers who came to show their work in detail, not in slogans.
Alex Mashrabov
Co-founder & CEO, Higgsfield
Diogo Almeida
Co-Inventor of ChatGPT & Co-founder, TypeSafe AI
Caitlin Colgrove
Founder & CTO, Hex
DJ Patil
Former U.S Chief Data Scientist
Barr Moses
CEO & Co-founder, Monte Carlo
Ten years of sessions from the archive. The clearest brief for a proposal is the talks that already worked.
Browse all talks
Why are some AI applications too good to be true and others total bunk? What is the reason behind the massive gap between over-promise and under-deliver? Let's talk about LLM history, mystery, and optimization to figure out where the f**k is all the automation.
The playbook for building a startup hasn't changed much in 20 years - until now. Fewer people, different skills, and a completely different sense of what's possible. A new generation of founders is shipping products with smaller teams, moving faster, and using AI across every function. Join a group of AI-native founders for a candid look at how they're actually using AI inside their own companies - in their workflows, their code, their ops - and what the shape of their teams looks like as a result.
While researchers may still opine about access to compute, most are hitting a new bottleneck: access to high-quality environments in which to post-train agents. Expanding agent capabilities will depend on how we curate and utilize data, much of which exists only in the minds of human experts today. This panel will explore the challenge of building environments, including the need to improve infrastructure reliability, the obstacles posed by subjective domains, and the challenge of achieving high variety. We’ll also discuss future opportunities, such as environments for long-horizon reasoning and continual learning, and what it will take to evolve environment design from a bespoke craft into a more automated, scalable discipline that consistently drives generalization.
Practitioners from the companies at the frontier of inference and model development sit down to unpack what application developers need to understand right now: why fine-tuning is quietly resurging under the name "RL," how smart teams are compressing inference costs by shaping smaller specialized models, who owns the model routing problem, and why inference capacity is structurally behind demand — possibly for years.
The lakehouse architecture solved a real problem — but AI has rewritten the requirements. Latency, compute separation, vector workloads, real-time inference pipelines: the demands look nothing like they did five years ago. Three engineers who helped build the modern data stack join us to debate what's actually changing under the hood, what's being thrown out, and what the next five years look like for the teams who have to ship on top of it.
In a recent New York Times op-ed, Mozilla CTO Raffi Krikorian warned that the informal detente that kept the internet secure — where writing software and finding vulnerabilities were equally hard — is over. Raffi sits down with Pete Soderling to go deeper: what the Mythos moment means for builders right now, how teams should think about AI as co-author rather than tool, the economics of open source in a world of expensive foundation models, and whether open models can offer a credible alternative to the frontier labs. Read Raffi's NYT essay: https://www.nytimes.com/2026/04/15/opinion/mythos-open-souce-internet.html
Proposals are read by the track hosts themselves. Tell us what you built and what you learned — the work matters more than having a previous speaking record.
The next edition is programmed by practitioners. Tell us what you built and what you learned.
Apply to be a speaker
