How Production AI Agents Are Actually Built | Mastra at Future Frontend 2026
AI is advancing across hardware, models, and inference, but Alex Booker and Tony Kovanen make the case that the biggest opportunities of the next decade sit at the application layer, in the harness and UX built around the model.
Recorded at Future Frontend 2026 in Helsinki, this talk uses Mastra as a lens for how production agents are actually built. Alex, who leads developer experience and education at Mastra, starts from a single weather agent and builds outward: model-agnostic setup, traces, evals and scorers, and the line between an agent and a workflow. Tony, a founding engineer at Mastra and co-creator of Next.js, then carries the demo to the frontend, streaming agent output into a live app with tool approval, human-in-the-loop suspension, and client-side tools.
You'll see observational memory rewrite an agent's context window in real time, watch harnesses extend past coding into entirely new disciplines, and get a tour of real agents in production at companies including Factorial, Sanity, Indeed, and Marsh. The session closes with an audience Q&A on context limits, agent security, and where personal agents go next.
📚 MASTRA RESOURCES
Mastra: https://mastra.ai
Mastra on X: https://x.com/mastra_ai
Mastra Discord: https://mastra.ai/community/discord
Mastra GitHub: https://github.com/mastra-ai
Learn Mastra in the world's first MCP-Based Course: https://mastra.ai/course
Principles of Building AI Agents (Book): https://mastra.ai/books/principles-of...
Patterns for Building AI Agents (New Book): https://mastra.ai/books/patterns-of-b...
WHAT IS MASTRA?
Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development—from prototype to production. You can integrate it with frontend and backend stacks (e.g., React, Next.js, Node) or run agents as standalone services. If you're a JavaScript or TypeScript developer looking to build an agentic or AI-powered product without starting from first principles, Mastra provides the scaffolding, tools, and integrations to accelerate that process.
0:00 Introduction
1:02 What is Mastra
5:20 Building your first agent
10:43 Agents vs workflows
12:49 Traces, evals, and scorers
15:01 Inside the Mastra server
17:06 Memory and context engineering
18:21 Observational memory in action
22:18 Reflections and prompt caching
24:17 Harnesses beyond coding agents
27:03 Agent use cases in production
29:46 Building an internal workshop agent
33:28 Streaming agents to the frontend
39:36 Client-side tools
41:46 Audience Q&A