Is Grok 3 the new best agent LLM?

Plus: long-term memory for agents, a roundup of our readers' most valued content, and more

Author

Michael Cunningham
February 20, 2025

In partnership with

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

I had a ton of fun on marketing guru Emanuel Rose’s podcast discussing the rise of AI agents and how anyone can build them, even without coding experience. If anyone thought that I was secretly an AI agent, sorry to disappoint you.

In today’s issue…

  • Musk’s xAI launches cutting-edge Grok 3 model

  • Long-term memory for AI agents

  • LangChain’s new agent marketplace

  • Can agents earn $1 million as freelance software engineers?

…and more

🔥 YOUR FAVORITES OF 2024

With Building AI Agents growing rapidly and nearly half of our readership having joined since January, I wanted to present a roundup of the 5 items our readers found most useful last year—if you’re a newcomer, I hope you do too!

📰 NEWS

Source: xAI

Elon Musk’s xAI released its much-hyped Grok 3 model on Monday, reporting results that surpassed other foundation models on multiple benchmarks and leaderboards. According to the company, Grok 3 and Grok 3 mini will be available for agent builders to use via its API in the next few weeks, featuring critical abilities such as tool use, code execution, and unspecified “advanced agent capabilities”.

The authentication provider’s new Connected Apps platform enables apps to securely give access permission to AI agents which can then access it using standard OAuth protocols, making any app accessible to the new agentic web.

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

🤝 WITH ARTISAN

Hire an AI BDR to Automate Your LinkedIn Outreach

Sales reps are wasting time on manual LinkedIn outreach. Our AI BDR Ava fully automates personalized LinkedIn outreach using your team’s profiles—getting you leads on autopilot.

She operates within the Artisan platform, which consolidates every tool you need for outbound:

  • 300M+ High-Quality B2B Prospects

  • Automated Lead Enrichment With 10+ Data Sources Included

  • Full Email Deliverability Management

  • Personalization Waterfall using LinkedIn, Twitter, Web Scraping & More

🛠️ USEFUL STUFF

Source: LangChain

LangMem SDK is a new library by LangChain that provides all three major types of memory for AI agents: semantic, episodic, and procedural. It integrates natively with LangGraph but works with any agent framework.

Startup Enso has partnered with LangChain to launch an AI agent marketplace where agent builders can easily monetize their agents, building them in LangGraph and launching them to the marketplace with just a few clicks.

💡 ANALYSIS

Source: Stytch

A blog post by Stytch demonstrating that different agents—OpenAI’s Operator, Anthropic’s Computer Use, and BrowserBase’s Open Operator—are blocked by different websites, with nearly every site being accessible to at least one.

Agentic technology could usher in a wave of automated cyberattacks, and cybersecurity experts are warning that the world is not prepared.

This piece is a good follow-on to Monday’s Building AI Agents spotlight on the displacement of robotic process automation (RPA) by AI agents, providing a effective playbook for enterprises seeking to move beyond RPA to agentic AI.

🧪 RESEARCH

Source: arXiv

This paper introduces a benchmark of real-world freelance software engineering tasks, collectively worth $1 million, and benchmark SWE agents’ performance on them.

Google advances the AI agents for science paradigm with a new multi-agent system designed to act as a researcher’s copilot. The science agent proposed multiple scientific hypotheses which were successfully validated in the laboratory.

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