Production AI patterns from teams who've already shipped.

Details coming soon.

Most engineering teams are stuck treating AI as autocomplete. Engineers have GitHub Copilot installed (or Claude or Cursor or whatever), they're generating snippets faster, but leaders can't connect usage to business outcomes—and developers are shipping code they don't fully understand.

Details coming soon.

AI adoption in software development is nearly universal, yet the outcomes for teams are highly variable. Why do some organizations see massive productivity gains while others see their delivery stability crash? DORA’s latest research provides a key insight: AI acts as an amplifier.

As organizations move from single-cloud setups to hybrid and multi-cloud strategies, they are under pressure to build data platforms that are both globally available and AI-ready.

Modern AI applications struggle not because of a lack of models, but because meaning is implicit, fragmented, and brittle. In this talk, we’ll explore how making semantics explicit (using ontologies and knowledge graphs) changes how we design, build, and operate AI systems.

ML training pipelines treat data as static. Teams spend weeks preprocessing datasets into WebDataset or TFRecords, and when they want to experiment with curriculum learning or data mixing, they reprocess everything from scratch.

Career progression in engineering has traditionally followed a predictable path: junior tasks teach fundamentals, mid-level work builds judgment, senior roles require synthesis across systems.

Early AI integrations often start small: wrap an inference API, add a prompt, ship a feature. At Zoox, that approach grew into Cortex, a production AI gateway supporting multiple model providers, multiple modalities, and agentic workflows with dozens of tools, serving over 100 internal clients.

We don't need LLMs to write new code. We need them to clean up the mess we already made.

In mature organizations, we have to maintain and migrate the existing codebase. Engineers are constantly balancing new feature development with endless software upkeep.

Your agent has 100k tokens of context. It still forgets what you told it two messages ago.

Coding models are helping software developers move even faster than ever before, but weirdly, they’re not keeping up with our fast progress. The models that power code generation are often based on months to years old snapshots of open source code.

AI agents are rapidly changing how users interact with software, yet most agentic systems today operate with little to no intelligent memory, relying instead on brittle context-window heuristics or short-term state.

Search is central to many AI systems. Everyone is building RAG and agents right now, but few are building reliable retrieval systems.