@gregisenberg X analytics

THE VIBE MANUFACTURING PLAYBOOK 1. Measure everything. Digital calipers for small parts, a phone LiDAR app like Polycam to scan a space or an object. Bad measurements are the #1 reason first builds fail 2. Have a frontier model write the design as code. Describe the part to a model like Claude Opus 5.5 or Astra, and have it write the design in a codebased CAD tool like OpenSCAD or CadQuery. Zoo's Text to CAD does it in one step! 3. Use real CAD files. The text to 3D models on Hugging Face make shapes that look right, but factories need precise files: STEP for machined parts, DXF for lasercut metal, STL for 3D printing 4. Make sure it can be made. Ask the AI to check the design against how factories actually work: how tight metal can bend, how small a hole can be, how much wiggle room parts need etc... 5. Prototype in plastic first. A $300 to $600 home 3D printer like a Bambu Lab lets you test the fit overnight before paying for metal (really fun to play with too!) 6. Pay an engin

It's kind of wild how many new primitives AI has given us in 3 years: - Local models (2023: Llama, Ollama, LM Studio) - Voice cloning (2023: ElevenLabs) - Text to CAD (2023: Zoo) - Text to video (2024: Sora, Veo, Runway) - Real-time voice (2024: Gemini Live, ChatGPT Voice) - Memory (2024: ChatGPT) - Reasoning models (2024: OpenAI o1, DeepSeek R1) - Vibe coding (2024: Bolt, Lovable, v0) - Computer use (2024: Claude, OpenAI Operator) - Connectors (2024: MCP) - Deep research (2024: Gemini, Perplexity, OpenAI Deep Research) - Agents (2025: Claude Code, Codex, Manus) - Image editing (2025: GPT-4o images, Nano Banana) - Agent payments (2025: x402, AP2) - Skills (2025: Claude Skills) - Personal agents (2026: Grokbot, Muse, Instinct)... we are here now. Each one of these is a starting point for hundreds of companies. The most interesting ones I'm seeing combine 2 or 3 of them with a specific niche!