2026 Schedule
Preliminary Schedule. The schedule is subject to change.
Monday, June 1st, 2026
Badge Pick-up (Stone Lobby) & Breakfast (Ziskind Lounge) - 1st Floor
Conference Introduction and Keynote:
Keeping ChatGPT Fast as AI Development Accelerates
Break - Snacks & Coffee Available in Ziskind Lounge
Beyond Prompting: Context Engineering for Production-Grade AI
If you're building production-grade AI apps, you may know an uncomfortable truth: reliable model outputs require far more than clever prompting.
Zero Trust Agent Systems that Pass Audits and Still Ship
Most agentic AI demos assume a greenfield environment. In a real enterprise, agents run inside strict boundaries where security, compliance, and incident response are non-negotiable.
From Prompt Hacking to Architectural Determinism: Engineering Reliable GenAI Systems
As generative AI moves from experimental novelties to mission-critical enterprise applications, the limitations of prompt engineering have become a significant bottleneck.
From Demo to Production: Why Agentic AI Systems Fail—and How to Fix Them
Most agentic AI systems look compelling in a demo. The agent reasons, calls tools, generates output. Then you try to run it in production, and the non-deterministic behavior of LLMs hits you fast.
SafeChat: Building AI-Powered Safety Systems at Scale in a Real-Time Marketplace
Trust and safety challenges become exponentially harder at marketplace scale, especially when interactions happen in real time.
Break
Context Engineering at LinkedIn: How We Built an Organizational Context Layer for AI Agents with MCP
AI coding agents are powerful out of the box, but they don't know your company. They can't navigate your services, understand your frameworks, query your data systems, or follow your organizational processes.
From Consumers to Builders: Turning 200 of our Team into Agent Creators in 2 Weeks
Most organizations approach "AI democratization" by giving coding assistants to engineers and simplistic chatbots to business users. Yet, building sophisticated, high-value AI agents remains a bottleneck restricted to a few specialists.
Your Newest Employee Has Sudo Access: Blueprints to Stop AI Insider Threats
Deploying a tool-calling agent is effectively onboarding a new employee who operates at a million actions per minute with database access and zero concept of consequences.
Serving LLMs at Scale: The Hidden KV Cache Advantage
KV cache is the hidden lever behind inference cost and performance. It directly impacts GPU utilization, throughput, and Time to First Token.
Leverage AI to Protect Your Data Assets... from AI
Most PII leaks happen before your prompt ever hits the model — here's how to catch them at the pipeline level with deterministic ETLs and synthetic stand-ins.
From AI Agent Demo to Production: Automated Testing and Evaluation
90% of the enterprise agents are stuck in POC. We don't know when the agents are good enough to deploy without a systematic testing and evaluation process.
RIIIR: Rewrite It In *Idiomatic* Rust (with AI)
Rust's strict type system and efficient execution make it a popular choice for agentic development -- but agents fail to make best use of Rust's idioms.
Solving Context Bloat: Semantic Tool Routing in Multi-Server MCP Environments
Agentic systems adopting MCP face a scalability hurdle: managing interactions with numerous servers exposing dozens or hundreds of tools.
Inferencing for Enterprises
This presentation will cover what areas enterprises like JPMC consider to be most important when running inferencing at scale.
The Trust Lifecycle for Safe, Secure and Reliable AI Agents
Shipping an AI agent is easy; keeping it trustworthy in production is not. Point-in-time audits fall short as failures emerge and attacks evolve. Teams must assess agent behavior across risks before deployment—and learn from production.
Break
Architecting the Data Layer for AI Agents: From Transactional Systems to MCP and Semantic Models
Most conversations about production AI agents focus on the agent itself — the prompts, the orchestration, the framework. But the moment you put an agent in front of real enterprise data, a different problem dominates: the data layer wasn't designed for this consumer.
Beyond the Prototype: Scaling Framework Agnostic AI Agent Infrastructure with Ray
Building an AI agent is easier than ever. However, moving from a local notebook to a production-grade "Agent Engine" that serves large scale web services poses complex problems.
Opening the Black Box: Critical Explainability for AI Agent Tool Selection with Kiji Inspector
AI agents that autonomously select and invoke tools are becoming ubiquitous—yet their decision-making remains a black box. When an agent chooses to query a database instead of searching the web, what internal representations drive that choice?
Vibecoding your Own Multi-Agent Workstation
I miss spending hours tuning my IDE, my dotfiles, my neovim setup, my PS1 prompt, my tmux config. In 2023, my workstation was hyper-customized to my preferences and needs.Now that agents write all my code, that work has gone in the bin.
Klaviyo Customer Agent: Building Bi-Directional Conversational Commerce at Scale
This presentation explores how Klaviyo is extending its core marketing philosophy — delivering the right message to the right person at the right time on the right channel — into fully bi-directional customer conversations through Customer Agent.
Break - Snacks & Coffee Available in Ziskind Lounge
Prompt to Prod: Engineering an Autonomous SDLC at Scale
Engineering organizations are hitting a Productivity Paradox. AI tools are generating more code than ever, yet the time it takes to ship that code safely to production in large-scale environments remains stagnant.
Decision-Driven Evaluation for Generative AI
Evaluation should not just produce metrics, which may turn out to be “metrics of convenience”, but should directly inform product and engineering decisions.
From Models to Agents: Building Context-Aware Consumer AI at Scale at DoorDash
Consumer AI is rapidly evolving beyond one-shot predictions toward systems that reason, plan, and adapt in real time.
Beyond Sandboxes: Architecting Durable Runtimes for AI Agents
Agent infrastructure has focused on sandboxing: running tools safely. But sandboxing protects tool execution, not agent operation.
Unconference Session: Brainstorming with Peers
What is Unconference?Unconference sessions are a simple way to run productive meetings for 5 to 2000 or more people, and a powerful way to lead any kind of organization in everyday practice and extraordinary change.
Terrace Lounge
Attendee Social - Snacks & Beverages - Located in Ziskind Lounge
Tuesday, June 2nd, 2026
Badge Pick-up (Stone Lobby) & Breakfast (Ziskind Lounge) - 1st Floor
Conference Introduction and Keynote:
The Five Stages of AI Maturity in Engineering Organizations — Where and Why Teams Get Stuck
Break - Snacks & Coffee Available in Ziskind Lounge
AI First, Quality Always: Agentic SDLC Adoption Case Study
Every engineering organization is under pressure to go "AI First" — fast. But the rush to show AI-driven productivity creates a dangerous incentive: measure speed and volume instead of quality and trust.
Building GenAI Platform at DoorDash
When we started adopting LLMs across DoorDash, every team was implementing the same infrastructure: retry logic, fallback mechanisms, cost tracking, prompt versioning, and batch processing pipelines. Engineering time was wasted on repetitive plumbing work instead of building features.
Scaling LLM-Based Ranking Systems for Latency-Critical Search & Recommendation Workloads
Large Language Models are powerful — but deploying them in latency-critical ranking systems is a fundamentally different problem than building chat applications.
From Fab To Token - The State Of The Market
This talk delivers a data-driven summary on the physical and economic bottlenecks in the AI infrastructure market today. We will cover the diverging strategies between traditional hyperscalers and specialized "Neoclouds," supported by deep supply chain data.
Autonomous Marketing at Klaviyo: Architecting Composer and 1:1 Personalization at Scale
This talk covers the our vision of autonomous marketing at Klaviyo, and goes inside the engineering behind Klaviyo's autonomous marketing system — the architecture, tradeoffs, and infrastructure that power AI-driven customer experiences across nearly 200,000 brands, 7.3B+ consumer
Break
The Agent Harness: Control Planes, Invariants, and Approval Boundaries for Production AI Agents
AI agents can look autonomous, but what keeps them reliable in production is the harness around the model.
From Hype to Habit: Conquering AI Adoption at Zoox
Most organizations have purchased AI developer tools. Most are disappointed with adoption. The problem usually isn't the tool, it's that we treat AI adoption like a software rollout when it's actually a behavior change problem.
Building Reusable Evaluation Frameworks for Agentic AI Products
This talk covers methods of evaluating AI Agents, with an example of how we built evaluation frameworks for a user-facing AI Agent system that has been in production for almost two years.
Multi-Agent Patterns from Spotify’s AI Powered Advertising Platform
Spotify's advertising platform spans audience targeting, ad creative generation, campaign goal resolution, budget optimization, and ad campaign recommendations -- workflows that touch different data sources, APIs, and business rules.
Building the Context Engine AI Agents Need
Every AI coding tool can generate code. Very few can generate the right code for your organization, because they're missing context.
Adaptive Recommenders in the Real World: Inference, Evals, and System Design
Modern personalization systems are shifting from hand-tuned heuristics to AI-native architectures, but building an adaptive recommendation engine in production—one that continuously learns, evolves, and delivers measurable business value—requires far more than deploying a model.
You Can't Trust Your AI Bill. And Neither Can I.
Cost, chaos, and what engineering teams get wrong about building on AI
Actors for Agents: Patterns for Production AI
We started with the obvious stack: a prompt-chain framework, some glue code, a queue for the slow parts. It demoed beautifully.
Making AI Reliable in the SDLC: The Role of Deterministic Context at Scale
Most enterprise AI initiatives stall not because of the model, but because of missing context. In large and complex systems, AI without visibility into architecture, dependencies, and constraints produces inconsistent outputs, erodes developer trust, and creates rework at scale.
The MCP Megalith: Surviving the 1,000-Tool Explosion and Auth Fragmentation
A year ago, remote MCP servers were largely experimental — alpha builds exposing a handful of tool operations. Today the ecosystem looks very different.
Break
Decision Models in Agentic Architectures: From Production to Agent Skills
In agentic architectures, not every step can be probabilistic. When an agent needs to decide whether to approve a loan, which treatment protocol to apply to a patient, or whether to pay an insurance claim, the output needs to be deterministic and explainable.
From Natural Language to Trusted AI: A Hybrid Architecture for Safe, Accurate, and Context-Aware AI Query Generation
Modern enterprises want everyone, not just data engineers, to be able to ask questions of their data. But turning natural language into reliable, secure, and business-aware queries is far from simple.
What 25 Trillion Tokens Reveal About How Developers Actually Adopt AI Agents
Most AI coding tools are benchmarked on capability. But capability doesn't predict adoption. At Kilo Code, we've processed over 25 trillion tokens across 1.5M+ developers, giving us one of the clearest real-world datasets on how engineering teams actually integrate AI into their workflows.
Finetuning Your Embedding Model for Better Search: Learning from Agentic Search at Dell
Embedding models are a foundational component of modern ML systems, enabling clustering, ranking, and large-scale information retrieval.
Make Quality Executable: Verification Gates for Agentic Engineering at Scale
AI coding agents are changing the economics of software delivery: teams can now generate more code, faster than ever before. But while generation has accelerated, code review, architecture governance, and engineering standards have not scaled at the same pace.
Break - Snacks & Coffee Available in Ziskind Lounge
From Retrieval to Reasoning: Building Production-Ready Agentic AI Systems with Knowledge Graphs
The evolution of AI systems is shifting from simple retrieval to autonomous reasoning.
Batch Intelligence at Scale: Cost-Efficient Multi-Agent LLM Workflows with Built-In Resilience
Most teams building production LLM systems face a hard tradeoff: premium models with low latency or cost-efficient inference with unpredictable quality. This talk presents a third path, one validated at Walmart, the largest retailers in the world.
Harness Engineering for AI Agents: Metrics, Systems, and Control
Agents fail in interesting ways. They don't crash, they drift. They don't error, they confidently do the wrong thing. Building a harness means designing for that specific kind of wrongness.
Unconference Session: Brainstorming with Peers
What is Unconference?Unconference sessions are a simple way to run productive meetings for 5 to 2000 or more people, and a powerful way to lead any kind of organization in everyday practice and extraordinary change.
Terrace Lounge
Closing Reception - Snacks & Beverages - Located in Ziskind Lounge
Evening Keynote:
