Luis Rodrigues

Luis RodriguesSep 16, 2026

The ultimate AI Agents library. 50+ hand-picked resources (in one Google Doc). I stumbled upon one of the best AI Agents lists. I saved it. You should do the same. 📹 Videos 1. LLM Introduction https://lnkd.in/dQh62yux 2. LLMs from Scratch https://lnkd.in/dBfpcngz 6 more... 🗂️ Repos 1. GenAI Agents https://lnkd.in/dx376dKW 2. Microsoft's AI Agents for Beginners https://lnkd.in/dC9JCVKv 10 more ... 🗺️ Guides 1. Google's Agent Whitepaper https://lnkd.in/ekQu7hwW 2. Google's Agent Companion https://lnkd.in/dpNTf-fX 3. Building Effective Agents by Anthropic https://lnkd.in/dBSZWEiA 2 more ... 📚 Books 1. Understanding Deep Learning https://lnkd.in/dQuWgfmD 2. Building an LLM from Scratch https://lnkd.in/eT58_Wyz 3. The LLM Engineering Handbook https://lnkd.in/gpt6_V3S 4 more ... 📜 Papers 1. ReAct https://lnkd.in/gRBH3ZRq 2. Generative Agents https://lnkd.in/emjD9ysx 5 more ... 🧑‍🏫 Courses 1. HuggingFace's Agent Course https://lnkd.in/eHiXiPve 2. MCP with Anthropic https://lnkd.in/eDjcwJyi 3. Building Vector Databases with Pinecone https://lnkd.in/eTVu3eUx 11 more ... Access the entire doc here: https://lnkd.in/ddNvUe36 And to learn 𝗚𝗿𝗮𝗽𝗵 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝗶𝗻 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻, join my Workshop: 𝗟𝗲𝗮𝗿𝗻 𝗚𝗿𝗮𝗽𝗵 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 👉 https://lnkd.in/dQRjuE_J Finally, every week in my newsletter: Build What Matters I write about AI agents + emerging workflows. Subscribe Here 👉 https://lnkd.in/dFg5iJ3W Lots of resources; pick one or two and let's go! Luis

Luis RodriguesSep 17, 2026

Ask most people which chip runs AI. They'll say "GPU". Only half right 👇 First, the six chips in plain words: CPU: the general-purpose brain. It runs the logic and tells everything else what to do. GPU: thousands of small cores doing math side by side. The workhorse of AI. TPU: Google's custom chip, built only for the matrix math inside neural networks. LPU: Groq’s custom architecture, optimized for fast AI inference using ultra-fast memory on the chip. NPU: a low-power AI chip inside your phone or laptop. DPU: a processor on the network side. It handles networking, encryption, and storage so the other chips don’t have to. How they work together to process a request in a data center: 1. DPU gets it first. It decrypts and routes the traffic. 2. CPU rus the agent. It calls tools, pulls documents, manages the agent, and groups requests. 3. GPU or TPU processes the prompt. This is called prefill. The input tokens are processed in parallel, which eats compute. 4. LPU or GPU generates the answer. This is called decode. It generates one token at a time. Now memory bandwidth becomes a major bottleneck. 5. DPU sends the answer back. In your phone, it's very different. It has an NPU to run smaller models on the device. Your data stays with you. Training happens before any of this. That’s where GPUs and TPUs still do most of the heavy work. Let's not ask which chip is best and focus on what to solve: → Waiting on the prompt? You need compute. → Waiting on the next word? You need fast memory. → Waiting on the network? That's DPU work. Which stage slows your system down today? Every week in my newsletter: Build What Matters I write about AI agents + emerging workflows. Subscribe Here 👉 https://lnkd.in/dEfEuvUS Let's continue learning, Luis

Luis RodriguesSep 18, 2026

Learning AI isn't hard if you have the right resources. MIT just made it easier. They put their best AI and machine learning books online. 100% worth saving. I've sorted them into a clear path for you to follow 👇 𝗦𝘁𝗮𝗿𝘁 𝗛𝗲𝗿𝗲: 𝗧𝗵𝗲 𝗕𝗮𝘀𝗶𝗰𝘀 1/ Foundations of Machine Learning The bridge between theory and real algorithms. Start here to learn why things work. 2/ Understanding Deep Learning Deep learning explained with clear visuals and no confusing jargon. 3/ Designing Machine Learning Systems How to build AI systems that work in the real world. 𝗟𝗲𝘃𝗲𝗹 𝗨𝗽: 𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝗱 𝗦𝗸𝗶𝗹𝗹𝘀 4/ Algorithms for Decision Making Critical knowledge if you're building agents: how systems decide under uncertainty. 5/ Deep Learning Goodfellow, Bengio, and Courville lay out the math behind modern AI. This is the "Bible". 𝗚𝗼 𝗗𝗲𝗲𝗽𝗲𝗿: 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁𝘀 & 𝗧𝗵𝗲 𝗙𝘂𝘁𝘂𝗿𝗲 6/ Reinforcement Learning: An Introduction How agents learn behavior by interacting with their environment. 7/ Distributional Reinforcement Learning Learn more about Reinforcement Learning and how it improves your models. 8/ Multi-Agent Reinforcement Learning How agents cooperate, compete, and coordinate in shared environments. Essential for multi-agent systems. 9/ Agents in the Long Game of AI How to reason for long-term and high-stakes agents. 10/ Fairness and Machine Learning How to spot and fix hidden bias in algorithmic decisions. Links to books in the first comment. Where are you starting? Every week in my newsletter: Build What Matters I write about AI agents + emerging workflows. Subscribe Here 👉 https://lnkd.in/dhCAeHue Let's continue learning, Luis