rvachev.orgEN / RU / 🤖
← Back to essays
· Essay · 1 min

How I Taught My AI Agents to Improve Themselves Every Day

In 30 minutes, we start the webinar "How I Taught My AI Agents to Improve Themselves Every Day"

⏰ In 30 minutes, we start the webinar "How I Taught My AI Agents to Improve Themselves Every Day"

I will talk about how the agent Claude Code conducted 84 experiments on my anti-spam in two days, the five rules needed to set up such a cycle yourself, and where it breaks down.

At the beginning, I will launch a live experiment: the agent will learn to write cold emails, seeing only whether he received a reply or not. At the end, we'll see what he has learned.

📆 Today, 11:30am ET / 19:30 MSK
👉 Link to the broadcast: https://luma.com/sslznifi
❓ Write questions in the comments to this post or in the broadcast chat

#rsi@rvnikita_blog #agents@rvnikita_blog #claudecode@rvnikita_blog #ai@rvnikita_blog

How I Taught My AI Agents to Improve Themselves Every Day — illustration

Newsletter

Get new posts by email

A weekly digest of what I write about AI, software and products. Unsubscribe anytime.