DeepLearning.AI

This week, in The Batch, Andrew Ng advises scaling down your AI projects so you can practice building AI apps even when you have only a short block of time. Plus: 🏗️ Amazon plans AI cluster in Indiana 🕶️ Meta details Aria 2 smart glasses 🌀 AI weather prediction gains traction 🧠 Chain of thought might leave gaps Read The Batch: https://hubs.la/Q03vJZ9f0

🚀 AI for Good in Action! We’re proud to be partners of the AI for Changemakers Program, supporting nonprofits in building AI-driven solutions for world’s most pressing problems, including global education. Over the past weeks, participants of the program - mission-driven organizations have developed real, impactful AI solutions, and now, it’s time to take the next step. On July 23, at AI for Education: Matching Day!, these nonprofits and social impact organizations will present their projects to connect with tech partners ready to help bring them to life. 💡 Why this matters: - 30 AI solutions tackling global education challenges - A chance for tech companies to contribute through pro bono partnerships - A global event with sessions across Asia-Pacific, EMEA, and the Americas 🔗Join this free, live online event and help match technology with impact: https://hubs.la/Q03vv0VJ0

To supercharge its AI push without triggering a government review that would come with acquiring other companies, Meta bought a 49 percent non-voting stake in data-labeling firm Scale AI for $14.3 billion and hired its founder and CEO, Alexandr Wang, and key staff. Wang will head a Meta research group that aims to develop superintelligence, while Scale AI gains capital to expand and reward shareholders, doubling its valuation to roughly $28 billion. Learn more in The Batch: https://hubs.la/Q03vv1sK0

📈 In business, your insight is only as valuable as the action it inspires. "Data Storytelling," the final course in the Data Analytics Professional Certificate teaches you how to structure your analysis into compelling, actionable narratives, tailored to the people making decisions. Learn to shape data into stories that drive change. Enroll today: https://hubs.la/Q03vtZhj0

ACP, this is incredible, this well presented course and the content, it is remarkable, thanks DeepLearning.AI, BeeAI, The Linux Foundation, Nicholas Renotte, Sandi Besen

🧠 Join the 10,000+ developers who already supercharged their #LLM skills with Reinforcement Fine-tuning (#RFT) — and it's free! 🧠 Reinforcement Fine-Tuning and #GRPO are fast becoming popular techniques to teach LLMs how to reason. We teamed up with DeepLearning.AI to build the definitive starter course for RFT including everything you need to know to become and GRPO pro! ✅ 1-hour #FREE short course   ✅ Hands-on labs: write #reward functions & train with GRPO and RFT   ✅ Learn how to train an LLM to master Wordle and other complex tasks   ✅ See exactly when RFT beats supervised fine-tuning (#SFT) ✅ Led by the researchers who built the first managed RFT platform Join the 10,000+ learners who’ve already leveled-up their model-tuning skills. Enroll free today and start shipping smarter #AI! 👇 Register for the free course: https://lnkd.in/g9z-wcUX And check out our AMA with the course creators: https://lnkd.in/gE9UuKsq

Researchers from Stanford and Princeton fine-tuned a language model to detect racial discrimination buried in millions of historical property deeds in Santa Clara County, California. These clauses, once used to bar people of color from owning or occupying certain homes, remain in documents despite being outlawed decades ago. The AI model analyzed over 5.2 million pages in just six days, identifying 24,500 lots with such clauses, work that would have taken a human nearly 10 years. Read our summary of the research in The Batch: https://hubs.la/Q03vf4HJ0

Bit of a dream come true with this one 🎉! We just released a couse on Deeplearning.AI 🤖 🦾 🚀 around all the amazing stuff you can do with ACP! In it you'll learn: 1. The differences between MCP vs ACP vs A2A 2. How to build ACP compliant agents 3. How to run pre-setup ACP agents like gpt-researcher, aider and more 4. Getting it all to work together as an agent system Thanks to Hawraa, Andrew and the team for the backing on this was an amazing project to get together. And a huge shoutout to my amazing co-lecturer Sandi, you smashed it. This was months of planning, coding, editing and recording to get this out but I'm super proud of the final product. Go check it out, link in the comments! P.s. I still have the photo in my office 😉

This week, in The Batch, Andrew Ng discusses a U.S. court ruling that training LLMs on copyrighted books is fair use. Plus: 💰 Meta invests $14.3 B in Scale AI 🧬 Biomni AI agent spans life-science research ⚠️ Top CEOs flag AI job cuts ⚡ Ternary BitNet beats most 2B models Read The Batch: https://hubs.la/Q03v2SxG0

New Course: ACP: Agent Communication Protocol Learn to build agents that communicate and collaborate across different frameworks using ACP in this short course built with IBM Research’s BeeAI, and taught by Sandi Besen, AI Research Engineer & Ecosystem Lead at IBM, and Nicholas Renotte, Head of AI Developer Advocacy at IBM. Building a multi-agent system with agents built or used by different teams and organizations can become challenging. You may need to write custom integrations each time a team updates their agent design or changes their choice of agentic orchestration framework. The Agent Communication Protocol (ACP) is an open protocol that addresses this challenge by standardizing how agents communicate, using a unified RESTful interface that works across frameworks. In this protocol, you host an agent inside an ACP server, which handles requests from an ACP client and passes them to the appropriate agent. Using a standardized client-server interface allows multiple teams to reuse agents across projects. It also makes it easier to switch between frameworks, replace an agent with a new version, or update a multi-agent system without refactoring the entire system. In this course, you’ll learn to connect agents through ACP. You’ll understand the lifecycle of an ACP Agent and how it compares to other protocols, such as MCP (Model Context Protocol) and A2A (Agent-to-Agent). You’ll build ACP-compliant agents and implement both sequential and hierarchical workflows of multiple agents collaborating using ACP. Through hands-on exercises, you’ll build: - A RAG agent with CrewAI and wrap it inside an ACP server. - An ACP Client to make calls to the ACP server you created. - A sequential workflow that chains an ACP server, created with Smolagents, to the RAG agent. - A hierarchical workflow using a router agent that transforms user queries into tasks, delegated to agents available through ACP servers. - An agent that uses MCP to access tools and ACP to communicate with other agents. You’ll finish up by importing your ACP agents into the BeeAI platform, an open-source registry for discovering and sharing agents. ACP enables collaboration between agents across teams and organizations. By the end of this course, you’ll be able to build ACP agents and workflows that communicate and collaborate regardless of framework. Please sign up here: https://lnkd.in/g4gES9CF