πŸ—πŸš’πŸš€ smol agents --> DeepResearch


​

Hey, AIM community!

​

Next Wednesday, join us to tackle DeepSeek-R1! We'll also train our own reasoning model with Unsloth while we're at it! Additionally, we'll dig into what we know about how the model was trained, and how it was used to distill Qwen and Llama models.

​


Last week, we explored Hugging Face's smolagents library and we went Deep on their new open-source DeepResearch implementation. What makes the library "smol?" How is it better than competitors? How did the HF team recreate OpenAI's Deep Research tool within days? We covered it all, from concepts to code!

🧰 Resources


πŸ”­ Coming Up!

COCONUT: Chain of Continuous Thought

Reasoning means doing things for a reason; thinking them through, step-by-step. Running different Chains of Thought (CoT) is a critial part of this process. Exactly how we create and leverage these CoTs can differ, and COCONUT leverages latent space.

Cursor: An AI Engineer’s Guide

Want to become an AI Engineer or leader? Start here to learn to build, ship, and share with a best-practice stack, including Cursor, uv, Chainlit, Docker, Hugging Face, and OpenAI. The Cursor of today is the start of the engineer of the future. Check it out.


🌐 Around the Community!

πŸ’‘ Transformation Spotlight: Akash Shetty! Learn how he went from generic software developer to full-stack engineer who has just launched an AI-poweredGovTech startup. Read more about his story.

video preview​

πŸ€“ See what the community is building, shipping, and sharing this week. Join us in the Lounge every Monday at 9 AM PT for some accountability!

​

Want to join the AIM community? Hop into Discord and share your intro!


​

πŸ–ΌοΈ Meme of the Week


🌟 Want to start building, shipping, and sharing, but not sure how? Check out our LLM Foundations - a 5-day email-based course.

​

Keep building πŸ—οΈ shipping 🚒 and sharing πŸš€,

​

​Dr. Greg, The Wiz, Seraacha, and Lusk​
​AI Makerspace​

​
​Unsubscribe Β· Preferences​

The LLM Edge

Read more from The LLM Edge
RAG: The 2025 Best Practice Stack

Hey, AIM community! Tomorrow, we'll cover Enterprise Agents with OpenAI! What does the agents SDK look like from OpenAI? How does it build on previous work they've done? Are they officially in the end-to-end platform game competing with orchestration frameworks like LangChain, LlamaIndex, CrewAI, and others? Join us live to find out! Last week, we discussed RAG: The 2025 Best-Practice Stack This is the year of Practical RAG, and we kicked it off by unpacking the Minimum Viable...

DeepSeek Week

Hey, AIM community! On Wednesday, we'll cover the infra stack that we recommend for RAG in 2025. Then, we'll build, ship, and share a best-practice RAG app. We'll also discuss important production tradeoffs and implications that you should consider before and after deployment when going from zero to production RAG! Last week, we discussed the latest open-source repo drops from DeepSeek Week, and we covered how they're being used as a new best-practice way to do inference on MoE models via...

Optimization of LLMs

Hey, AIM community! Next Wednesday, we begin a new series on Optimization of LLMs! We'll tackle an important topic from first principles: building and optimizing LLMs before they make it to production. What are the essential concepts and code that underlie the technology, from loss functions and gradient descent to LSTMs, RLHF, and GRPO? Join us to kick off our new series - which we will continue monthly - about Optimization of LLMs. Last week, we put PydanticAI to the test! πŸš€ The team behind...