1. The Frontier Giants • Gemini: Access 1.5B tokens/day on Gemini 1.5 Flash/Pro. That is an astronomical amount of context for RAG and long-document analysis. • OpenAI: Their “Data Sharing” program offers 250k/2.5M tokens daily. • xAI Grok: Spend just $5 and unlock $150/month in free credits. • Amazon AWS: New users get $100 credit for 6 months, providing access to 200+ models including Opus 4.7 and GPT 5.1. 2. Speed & Open-Source Powerhouses • Groq: The king of inference speed. Access Llama 3.3-70b and Qwen3-32b at speeds that feel like magic—completely free. • Mistral: Their Experimental Program offers a massive 1B free tokens per month. • Nvidia: Use the Nemotron suite via their developer playground for high-performance base models. 3. The Aggregators & Community Hubs • Hugging Face: The "GitHub of AI" provides a Free Serverless Inference API for thousands of models (Llama, Stable Diffusion, Whisper). No credit card required. • OpenRouter: Access 50+ models with unlimited usage tiers for experimentation. • Deepinfra: Get 1M tokens/day on Llama/Mistral models just for signing up with an email. 4. Specialized & Niche Access • Cohere: Their Trial API gives 1,000 calls/month for the best-in-class Rerank v3 and multilingual Aya models. • Lepton AI: $10 free credit on signup to test Llama and Gemma models in a streamlined playground. So, what are building today?

I did a quick search here, but didn't see many posts that even mention the French AI/LLM "Mistral" (https://mistral.ai/)? What are some of the community's opinions or experiences with this? Even YT doesn't have much on it. Upside: Frontier model, open weights, options to self host, token costs (less than the big three, more than the Chinese AIs), 25% cheaper basic paid subscription ($15US instead of $20). Downside, it's in the EU, it's French (lol) and there's the whole digital wallet thing and EU nanny state mentality. I've been looking at it's features with a few chats this week and I wonder if it's a viable alternative. This is about the best video I've found on it so far: "The AI Model Nobody's Talking About — Mistral 3.5 & The Sovereignty Question" https://www.youtube.com/watch?v=LkbREGlNX8Y

Here’s something that’s been sitting on my mind. We’re in a moment where a lot of people are saying learning to code is no longer necessary. That AI can just write the code for you. And on the surface, that argument sounds reasonable. But I think it misses something fundamental about how software actually works. The abstraction stack has always looked like this: Natural Language → [AI translates] → High-Level Code (Python, JS) → [compiler] → Assembly → [CPU] → Machine Language This is the same pattern software has always followed. We went from punching machine code, to assembly, to C, to Python. Every layer up was an abstraction that made the layer below more accessible. AI is simply the next one. You can now describe intent in plain English and get working code back. That’s powerful. That’s genuinely a shift. But here’s what hasn’t changed: The AI still produces code. That code still runs on software engineering principles, and neglecting those principles is just like pushing code that’s never been vetted. Same same, but different. If you can’t read what it produces, you can’t evaluate it, debug it, extend it, or know when it’s wrong. Think about it you wouldn’t trust a translator if you had no idea what language they were translating into. Same logic applies here. And here’s the honest truth: learning a programming language alone isn’t enough right now, because AI is doing that part better by the day. What actually matters is the engineering fundamentals. They teach you how to think in terms of logic, data flow, state, and structure. Those aren’t things AI removes from the equation they’re the things that help you direct AI well. So do we still need to learn to code? My take: software engineering fundamentals are non-negotiable. Understanding how code works, what a function does, what an API call is, how data moves through a system these matter more now, not less. You need to be a good reviewer of AI output. And knowing the fundamentals lets you govern what AI writes and how it writes it throughout the entire development process.

I found 3 interesting GitHub repos for Claude / Claude Code that might be useful if you’re building websites, UI, or frontend projects with AI. The idea is pretty simple: Instead of just telling Claude “make this better”, you give it a proper skill — basically a small set of rules, taste, and guidance for how to think about layout, spacing, typography, animations, and UI polish. Here are the 3 repos: 1. Impeccable https://github.com/pbakaus/impeccable This one feels like the most complete of the three. It helps Claude review and improve frontend work — things like layout, typography, spacing, colors, responsive issues, UX details, and overall polish. I think this is especially useful when you already have a page or component built, but it still feels a bit unfinished or “AI-generated”. 2. Taste Skill https://github.com/leonxlnx/taste-skill This one is more about helping AI stop making generic SaaS-looking designs. It pushes Claude toward better visual taste: better composition, better spacing, better motion, better style, and stronger design direction. I like this one because it’s not just technical. It tries to help the AI make better design decisions. 3. Emil Kowalski Skill https://github.com/emilkowalski/skill This one is more focused on design engineering and the small UI details that make a product feel polished. Think animations, micro-interactions, transitions, smooth UI behavior, and that “premium” feeling. In my opinion, this is best when the design is already decent, but you want to make it feel more refined. My takeaway: I wouldn’t install everything at once. I’d start with Taste Skill or Impeccable, test it on a real project, and only keep what actually improves your workflow. AI coding tools get much better when we don’t just give them prompts, but also give them structure, examples, and clear design rules.

1. Watch the intro video and introduce yourself in the intro post here 2. Start with The Foundation (free course). Concepts, folder architecture, prompting framework. Everything else builds on this. 3. Check in at the bottom of each lesson. Polls, discussion posts, other members working through the same stuff. Use them. 4. When you're ready to build real things, move to Implementation Playbooks (Level 2). When you're ready to build your own tools, Building Your Stack (Level 3). 5. Post your work. Ask questions. Help others when you can. What are you here to build?