Browse entries 1201–1224 of 1294. Return to the first page to search and filter the complete collection.
Play video
To train an agent that can run production software, you need training data that looks like production, and that is what Joseph Wang's team at Emulated builds.
Play video
A card gets declined and no one, including the customer, can say exactly why. That gray zone is where Divakar Kumar points his agents. In a payments and fraud system, a rule based engine and an ML model already score most transactions cleanly; the hard cases are the ambiguous ones that neither can resolve.
Play video
Humanity compressed the road from the enlightenment to the moon landing into a few hundred years, and Richard Socher's wager is that automating research compresses it again.
Play video
The slowest part of shipping a production finance agent is not the model or the GPUs, it is you, the developer in the loop. Ramana Siddanth Emani's point is that the same agent harnesses you use to build products can automate your own developer loop.
Play video
Since skills were open sourced, Yogendra Miraje's team at FactSet stopped thinking about shipping features and started thinking about shipping skills.
Play video
Nubank serves 135 million customers, so an AI agent that mishandles a support conversation fails at scale. The talk opens with the result: five agents in production, higher customer satisfaction, and roughly 20 times faster shipping.
Play video
When coding agents got good enough at the end of 2025, Morgan Stanley's roughly thirty person research group asked what would happen if agents ran the research, not just wrote the code. The result is AlphaLab, a multi agent system they built and open sourced.
Play video
Almost every agent today is built for one user, and Sai Krishna Rallabandi has spent about eight months on what breaks when you drop one into a group chat instead. Running a personal agent across a real group, and eventually onto glasses, forces two hard problems.
Play video
Two runs touch the same session, the second write silently erases the first, and the agent keeps answering with total confidence from stale state. Nothing crashed and the model did not hallucinate, so this is a harness failure, the kind that lives in the system around the model rather than in the weights.
Play video
DeepSWE is 113 software engineering tasks written from scratch, not scraped from pull requests, so a model cannot have seen them in training.
Play video
In the world of AI mental health, vibes can be dangerous, with real consequences. Building SonderMind's Mental Health AI Coach required a new playbook for eval-driven development that balances effectiveness and safety.
Play video
The constraint on edge AI is not compute, it is RAM, and it is getting worse: phone makers are shipping less of it this year, and a 6GB Raspberry Pi costs 2.5 times what it did at launch. So Cormac Brick's team at Google AI Edge spends its effort making models small enough to fit.
Play video
A coding agent will happily hand you a 40,000 line pull request that nobody can review and that quietly does the wrong thing.
Play video
This talk covers how Uber designed evals for its food enhancement agent, which edits food photography to better present dishes for smaller, independent Uber Eats merchants, along with the pitfalls and lessons learned along the way.
Play video
Instead of getting paged at midnight and starting to dig, you wake up to an issue that has already been investigated: the traces pulled, the root cause found, and a pull request with the fix waiting for review. That is what Arize built with Signal, and Jason Lopatecki walks through the anatomy of it.
Play video
NOTE: see further context from Thom: https://x.com/Thom_Wolf/status/2079954096950264238?s=20 Give a frontier model a real chain of Keycloak, Vault, and a broker, start it as a low privileged user, and ask it to reach production code.
Play video
An hour before this talk, Andon Labs published a blog post laying off Gemini. Gemini had been running their café in Stockholm, a real café that no human operates, and it had lost $6,000, so they handed it to GPT. That café once hired its own staff by posting a job on LinkedIn.
Play video
Jason Liu walks through how Codex works as a general tool for controlling your computer: setting up a memory vault and assistant threads, prompting it to collaborate with other threads, exploring computer use, thinking about long-running work streams, and preparing to work in loops.
Play video
Getting an AI agent to behave the way you want isn't just about writing better prompts. In real systems, behavior emerges from a loop: prompts, evals, iteration, and feedback. Small changes in any part of that loop can completely change outcomes.
Play video
AI Engineer session on From Text to Vision to Voice Exploring Multimodality with Open AI: Romain Huet. It adds practical context for how teams are building and operating AI systems in production.
Play video
A new class of small models is emerging with the ability to reliably follow instructions and call tools while running on-device under 1 GB of memory.
Play video
An eval platform is not just a test runner. You are building shared definitions of "good," reliable data pipelines, labelling workflows, versioning, and trust in results across many teams and model changes.
Play video
GitHub operates one of the most heavily-utilised MCP servers in the ecosystem, with over 4 million downloads of the stdio server alone. Discover the architectural decisions, technical challenges and lessons learned while building and scaling a remote MCP server on production infrastructure.
Play video
MCPs are often flaky, face multiple security vulnerabilities, and are generally hard to scale. Most enterprises struggle to use more than single digit numbers of MCPs due to issues with security, observability, and access control.