Daytona AI Builders

An event dedicated to exploring all things AI Engineering!

Event partners: AWS & Antimetal

Agenda

🕒 5:30 pm – 6:00 pm

Welcome Gathering and Opening Remarks

🎤 Marijan Cipcic, Principal Events Manager at Daytona

🕒 6:00 pm – 6:15 pm

Talk "Forkable Sandboxes for Long-Running Agents"

🎤 Muhammad Annas Hashmi, DevRel at Daytona

Outline:

Agents are starting to do real software work: installing dependencies, editing files, running tests, starting servers, and debugging over long sessions. But most AI systems still treat each run like a disposable prompt-response loop. That breaks down once the useful state lives not just in tokens, but in the actual machine.

This talk is about the runtime layer agents need to become reliable builders. I’ll show how forkable Daytona sandboxes let you pause at important moments in an agent run, branch the environment, and sample multiple continuations from the same machine state. For builders, this turns long-running agent workflows into something you can inspect, compare, recover, and eventually optimize.

🕒 6:15 pm – 6:30 pm

Talk "Multi-Agent Investment AI Assistant built on Amazon Bedrock Agentcore"

🎤 Sam Farber, Sr. Solutions Architect at AWS

Outline:

Investment research teams spend days manually gathering data across markets, risk metrics, sentiment indicators, financial statements, and alternative data sources before synthesizing findings into actionable reports. This talk will show how a multi-agent AI system can automate that workflow. A supervisor agent breaks down a research question, delegates to 10 domain-specific specialists running in parallel, and compiles their findings into a comprehensive investment report. The system maintains conversation memory across each agent, allowing researchers to have targeted conversations and follow-ups with each specialized agent.

🕒 6:30 pm – 6:45 pm

Talk "Building a World Model for Production Agents"

🎤 Brian Feng, Founding Engineer at Antimetal

Outline:

Production systems are increasingly complex, with critical knowledge scattered across code, infrastructure, telemetry, and people's heads.

In this talk, we'll explore how to build a persistent world model that gives AI agents a shared understanding of production. We'll break down four foundational layers: physical topology, temporal changes, causal relationships, and organizational context to show how they enable agents to investigate incidents, assess changes, and proactively manage production, building on what they've learned across tasks.

🕒 6:45 pm – 6:55 pm

Talk "Curing Omnichannel Amnesia with Twilio and n8n"

🎤 Kaelyn Chresfield, Developer Evangelist at Twilio

Outline:

How do you build AI agents that remember users across channels without overwhelming the context window? In this talk, we’ll use Twilio Conversations and n8n to build an AI agent that maintains user memory across SMS and voice. Through a live demo, we’ll explore how to capture, store, and retrieve user information to create seamless, personalized conversations across channels.

🕒 6:55 pm – 7:05 pm

Talk "Letting AI agents use credentials they never see"

🎤 Chase Erickson, Solutions Engineer at Infisical

Outline:

Infisical manages the secrets your Daytona agents depend on. The Daytona secret sync pushes them from Infisical to your Daytona organization and keeps them current, rotations included. Infisical's Agent Vault, a proxy, adds the credential to requests bound for the services you configure, so the agent never sees it. In this talk, we'll show how an agent in a Daytona sandbox calls an API without ever holding the secret.

🕒 7:05 pm - 8:30 pm

Networking

With pizzas and beverages

About event

This is dynamic gathering for AI enthusiasts, innovators, and professionals to collaborate, share ideas, and explore the latest advancements in artificial intelligence. Whether you're building AI products, researching cutting-edge algorithms, or simply passionate about the field, join us to connect, learn, and drive the future of AI forward.