Nvidia CEO reveals major agent push

Plus: LangChain's agent conference, how to build good agent evaluations, and more

Author

Michael Cunningham
January 09, 2025

In partnership with

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

Getting your terminology straight is important! Also, AGI = any new LLM that did slightly better on benchmarks than its predecessor

In today’s issue…

  • Nvidia bets big on AI agents

  • LangChain’s upcoming agent conference

  • How to build agent evaluations

  • Transforming industries with vertical agent SaaS

…and more

📰 NEWS

Jensen Huang at CES 2025 | Source: Nvidia

Nvidia CEO Jensen Huang’s keynote kicking off CES 2025 leaned heavily into agentic AI, declaring “The IT department of every company is going to become the ‘HR department’ of AI agents.” The GPU giant released the new Llama Nemotron family of agent-focused LLMs and a set of blueprints for building enterprise agent systems, and announced a collaboration with agent framework provider CrewAI to integrate its technology into Nvidia’s ecosystem.

The company behind the leading LLM orchestration framework will hold a conference devoted to agents in San Francisco in May, with tickets going on sale in mid-January for $499.

As part of its 2025 meeting, The Web Conference is holding a competition to design LLM-based agents to enhance the web experience with a total prize pool of $12,000.

Accenture and Nvidia further expanded their AI Refinery partnership, rolling out a new suite of tools to help enterprises build and integrate AI agents.

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

🤝 WITH WRITER

Writer RAG tool: build production-ready RAG apps in minutes

  • Writer RAG Tool: build production-ready RAG apps in minutes with simple API calls.

  • Knowledge Graph integration for intelligent data retrieval and AI-powered interactions.

  • Streamlined full-stack platform eliminates complex setups for scalable, accurate AI workflows.

🛠️ USEFUL STUFF

Source: YouTube

An in-depth video tutorial by LangChain showing how to build effective evaluations for agents and use them to improve agent performance.

An open alternative to Claude Computer Use, powered by E2B’s virtual Linux OS for agents and open-source LLMs.

A simple tutorial for implementing agentic RAG with IBM’s Granite 3.1 model for enterprise agents.

💡 ANALYSIS

Source: Wikipedia

A report focusing particularly on tools for enhancing agents such as function calling and data stores, with code examples throughout.

Chip Huyen’s deep dive into several crucial concepts in agent building such as tools, planning, and dealing with common failure modes.

A digest of Accenture’s Technology Vision 2025 report, which argues that enterprises of the future will be built around central “Cognitive Digital Brains”, where company data is accessed by armies of AI agents.

Five real-world case studies from major companies such as Johnson & Johnson and eBay, demonstrating that major companies are already putting agentic systems into production.

Though most of the hype around AI agents has focused around tech giants such as Microsoft and Salesforce, smaller robotic process automation (RPA) firms such as UiPath are rapidly pivoting into the space. In this piece, UiPath lays out their vision for agentic enterprise automation.

🧪 RESEARCH

Core architecture of an AI agent, from the paper below | Source: arXiv

This provocatively-titled paper gives an introduction to agentic systems and discusses their potential to displace traditional SaaS and transform specific verticals by automating their workflows.

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