Microsoft's agent course for beginners

Plus: easily build a web scraping agent without coding, how outdated infrastructure could hold agentic AI back, and more

Author

Michael Cunningham
March 06, 2025

Welcome back to Building AI Agents, your biweekly guide to everything new in the AI agent field!

Ten beers later, it experienced model collapse.

In today’s issue…

  • The new Replit Agent is here

  • Microsoft’s lesson series for new agent builders

  • Easily build a web scraping agent without coding

  • How IT infrastructure could hold agents back

…and more

📰 NEWS

Source: Replit

Replit Agent, which created a stir last year with its astonishing ability to autonomously code entire apps, is back with a new version.

AI-powered search startup Perplexity opened the waitlist for Comet, a “browser for agentic search”.

If you find Building AI Agents valuable, forward this email to a friend or colleague!

 🤝 PARTNER WITH US!

Do you have a high-quality product or service agent builders and business leaders would like to hear about?

Building AI Agents has…

👥 Thousands of readers

📈 Rapid audience growth

🙌 High engagement and satisfaction

If you’re interested in advertising with us, please reach out here!

🛠️ USEFUL STUFF

Source: Microsoft

A series of intro lessons on AI agents covering agent frameworks, tool use, design patterns, and more.

This simple tutorial shows how to use popular no-code framework n8n to obtain and process Google search results.

A library by LangChain enabling users to build swarm-style agents, which specialize in specific roles and complete tasks by intelligently handing them off to each other.

An e-book providing a deep dive into single- and multi-agent architectures, particularly for applications which require information retrieval at scale.

💡 ANALYSIS

Created by the author using Dall-E 3

The main barrier to transforming companies’ operations with AI agents will be outdated data systems, not the agents’ intrinsic capabilities, this piece argues

Slack’s Chief Marketing Officer predicts that employees will soon spend more time at work interacting with AI agents than with co-workers, consistent with NVIDIA CEO Jensen Huang’s vision of hundred-million-agent companies.

🧪 RESEARCH

Source: Wikipedia

This paper by Google researchers presents a new framework for agents to intelligently approach problems which can be used by any model.

Thanks for reading! Until next time, keep learning and building!

What did you think of today's issue?

Login or Subscribe to participate in polls.

If you have any specific feedback, just reply to this email—we’d love to hear from you

Follow us on X (Twitter), LinkedIn, and Instagram