Max Gfeller
Technical evangelist at Lovable
Stockholm, Sweden
Actions
Max Gfeller is a technical evangelist at Lovable and was previously the tech lead for AI at Sutro, where the team built AI infrastructure that generates production-grade backends from text prompts. He created OpenHarness, an open-source agent harness (600+ GitHub stars), and co-founded Storique.ai, an AI-powered personalized storybook platform that trains custom diffusion models on your photos. He has spoken at many conferences around the world, including LambdaConf, Techorama, Developer Week, AI Tech World, and many more.
Area of Expertise
Topics
Build your own Lovable
Coding agents can feel like magic: point one at a codebase, ask it to fix a bug, and it reads files, edits code, runs commands, and iterates until the task is done. But under the hood, the core ideas are surprisingly approachable.
In this talk, we’ll build up a small coding agent from scratch to understand how tools like Claude Code, Cursor, and Codex work. We’ll look at the agent harness: the application layer that talks to the model, manages context, executes tool calls, and keeps the work moving.
Starting with a simple conversation loop, we’ll add file and command access, use a browser to verify the agent’s work, and explore what happens when its context fills up. We’ll then look at subagents, project instructions, MCP, and reusable skills. Along the way, I’ll connect these building blocks to engineering examples from Lovable, including how feedback from the agent can help us improve its harness.
You’ll leave with a working mental model of coding agents and concrete patterns you can use to build your own, customize the tools you already use, and debug them when they go wrong.
This session is very much hands-on and can be delivered as a classical conference talk with live coding and/or live demonstration, or as a workshop in which attendees build their own coding agents.
Building apps for agents: MCP lessons from Lovable
MCP’s newer capabilities let us build richer interactions between assistants, applications, and people: server events, long-running tasks, requests for user input, and interactive MCP Apps.
In this demo-driven talk, we’ll explore what those capabilities make possible, from reacting to changing data to following background work and using an app inside a conversation. Drawing on lessons from Lovable, we’ll share what worked, what broke, and what makes these integrations reliable. You’ll leave with ideas for your own applications and a clear view of what’s supported today and what’s still emerging.
Talk · 40 min
What building millions of apps taught us about React
Generating a React page is one thing. Keeping it working through the next twenty changes is harder.
At Lovable, users have created more than 70 million projects. Drawing on lessons from our engineering teams, we’ll explore what building and evolving these frontends has taught us: choosing defaults such as TanStack Start, learning from recurring mistakes, and giving agents feedback from the running app. Through concrete examples, you’ll learn practical ways to structure React projects and check agent-generated changes without losing control of the result.
Talk · 40 min
Who maintains software when everyone can build it?
Building an app with AI can take an afternoon. Who keeps it secure and working six months later, especially when its creator doesn’t know how?
Drawing on lessons from Lovable, where users have created more than 70 million projects, we’ll follow an AI-generated app beyond its first deployment. We’ll explore secure defaults, automated checks, and repairing vulnerabilities without breaking what’s already live. You’ll leave with practical patterns for maintaining generated software, verifying automated fixes, and deciding when the owner needs to be involved.
Talk · 40 min
Making AI feel fast
An AI app can respond quickly and still leave people waiting. Streaming an answer doesn’t help much if the interface freezes or a follow-up message sits behind a long-running task.
Drawing on lessons from Lovable, we’ll improve an intentionally frustrating AI app and measure what changes. We’ll look at responsive interfaces, background work, and matching the model to the task, while letting users steer the agent before it finishes. You’ll leave with a practical approach to reducing unnecessary waits and keeping people involved while the work happens.
Talk · 40 min
AI beyond the chatbot
For a while, every AI feature seemed to arrive as another chatbot. But making people describe everything in a text box often creates more work for them.
Drawing on lessons from Lovable, we’ll take a familiar chatbot app and rethink its UX: turning answers into actions and interactive interfaces, connecting useful tools, and remembering what matters to the user. We’ll also look at when help should be proactive and when it belongs outside the app. You’ll leave with practical ways to make AI features more capable and a pleasure to use.
Talk · 40 min
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top