Hey everyone! Really looking forward to tonight’s session — we’re leveling up! Last week we broke down how AI actually works under the hood: LLMs, prompt engineering, and why the way you talk to AI changes everything. Tonight, we’re taking that foundation and running with it. Here’s what’s on deck: Context engineering — the secret sauce behind great AI output (hint: it’s not just about better prompts) Context windows & RAG — how AI pulls in fresh, relevant information instead of relying on what it was trained on Agent memory — how AI remembers what matters and how you can set up simple file-based memory systems yourself The big takeaway? When you build a better environment around AI, you get dramatically better results. Think of it like this — a great chef needs a great kitchen, not just a great recipe. Missed last week? No stress at all — I’ll kick things off with a quick recap so everyone’s on the same page. By the end of tonight, you’ll walk away knowing how modern AI tools actually work and how to start designing your own workflows to make AI sharper, more personal, and way more useful. See you there!

Today I'll be covering what machine learning is. Last week we covered the basics of what AI is in general and this week we'll be diving into machine learning specifically. Hop on if you want to listen or ask questions. See you then 🙂

Claude’s new model so far seems to be living up to its hype. At least inside claude code, it is faster and runs more efficiently than Opus 4.8. I haven’t of course run any personal evals on the model yet but I can notice some performance bump. Have you tried it? How was your experience?