buttercup.sh

Em dashes welcome — learn how to build AI agents

8 followers

Em dashes welcome — learn how to build AI agents

8 followers

Visit website

Engineering & Development

•

LLMs

•

AI Agents

Join the next few weeks and learn to build AI agents. Open source. Courses posted on Youtube and buttercup.sh. Learn how to build AI Agents. You may be interested in the high-level details building agents and software. It's important to know what is possible, so you can learn what to ask code agents to add it to your apps. By knowing what is available, you'll be able to build more robust apps and agents.

Free

Launch tags:Artificial Intelligence•GitHub•YouTube

Launch Team

Startup Metrics Dashboard

Maker

📌

Join us over the next few weeks to learn how to build AI agents. Take a look at the source code on github to learn how AI agent harnesses are built. We will be adding more topics and ideas every week. Starting mid-September 2026.

Report

2mo ago

Forum Threads

p/buttercup-sh-web-agent-harness

•

6d ago

Final Lesson: Evals

It was a lot of fun to talk with you all and building these lessons over the last two months. Thank you for joining me and here is the final lesson for building your own Agent: https://buttercup.sh/lessons/202...

p/buttercup-sh-web-agent-harness

•

13d ago

Lesson 5: vision - help your agent see what it built

Welcome back! In this lesson, you ll give your agent a pair of eyes. We ll add screenshots to the loop from lesson 3 so the model can read a layout, spot a button beyond the viewport, and compare the result with your request. A phone-sized screenshot costs about 333 input tokens, and a simple DOM check confirms when the page is ready. By the end, your agent will be able to look, improve, and check its own work.

p/buttercup-sh-web-agent-harness

•

20d ago

Lesson 4: agent communication - the agent that hires another agent

Lesson 4: agent communication

Build an agent that delegates work to agents written by other people. Publish an agent-card.json and a handler on Blocks.ai, accept tasks with blocks run, and call another agent from the handler with ctx.taskClient.sendMessage(). Lesson 3's turn counter becomes a 60-second wall-clock budget.

View all