Brewhubphl

The Master Series Introduction: Field Notes from a Coffee Shop Spaceship open.substack.com/pub/brewhub/p/…

"Franklin: a coffee-shop AI that treats neurodivergent customers as regulars" by BrewHubPHL #DEVCommunity dev.to/brewhubphl/fra…

"The Tri-State Stack: Why We Split Next.js, Python Agents, and Postgres Across Three Different Hosting Models" by BrewHubPHL #DEVCommunity dev.to/brewhubphl/the…

The Master Series Introduction: Field Notes from a Coffee Shop Spaceship open.substack.com/pub/brewhub/p/…

sketchpad_10.md: The Debt That Was Hiding Bugs open.substack.com/pub/brewhub/p/…

sketchpad_09.md: The Bug That Wasn't There open.substack.com/pub/brewhub/p/…

sketchpad_08.md: Instagram DMs at Scale open.substack.com/pub/brewhub/p/…

sketchpad_07.md: The Agent Got the Keys open.substack.com/pub/brewhub/p/…

@Chaos2Cured @ClaudeDevs @bughuntergeek @CommunityNotes I’ve ran 3 /worklfows and during all 3 Fable triggered itself

@CommunityNotes 1. This is marketing. It is not true. 2. People can’t ask about medicine, biology, chemistry, consciousness, psilocybin, LSD, or building AI. Also, not fractal math, fractal programming, or building autonomous AI. 3. I am tired of labs lying about “safety” when it is all for them to stop anyone from competing with them. Their guardrails are BS, and they know Fable is flopping, and they are worried about their IPO, so, back to point one. Just read it in circular fashion. If community notes is real, it is about time the labs were called on their BS. Sorry, i don’t have better sources to site. Test it. Test Fable. And everyone out there reading this… report every time you get shot down for a real question that shouldn’t be shut down. •

We’re rolling out changes to make Fable 5’s safeguards for frontier LLM development visible. Starting this week, flagged requests will visibly fall back to Opus 4.8—the same as our safeguards for cyber and bio. You will see this every time it happens. On the API, any flagged requests will return a reason for their refusal (coming to server-side fallback in the next few days). We wanted to deploy Fable 5 to our users quickly and safely. Visible safeguards can be probed, so they have to be robust, which takes time to get right. Invisible safeguards can be targeted more narrowly, allowing us to ship quickly with very few false positives. We went with invisible safeguards for this reason—and that was the wrong tradeoff. You should have visibility into the safeguards we have in place, and why. We’re sorry for not getting the balance right. Making the safeguards visible makes them easier to work around, so keeping them robust to jailbreaks will unfortunately mean more false positives while we improve the classifiers. We're also tuning our bio and cyber classifiers to trigger less often on harmless requests. We know this is frustrating and we’ll do our best to keep this period as short as possible. If you think a request has been mistakenly flagged: run /feedback in Claude Code, click thumbs-down on the fallback in Claude.ai or Cowork, or file the safeguard appeal form for API requests. Your reports help us tune these classifiers and we appreciate your feedback. support.claude.com/en/articles/82…

sketchpad_06.md: Sweeping the Floor open.substack.com/pub/brewhub/p/…

If you’re building something and the values feel like marketing until they hurt a little… you might be doing it right. What’s something you’ve hard-coded into your system that protects the culture?

BrewHub opens in 6 months. The baristas already know the rules. So do I.

We built our ops system so that no one — not even me, the owner — can do any operational work without clocking in. No “do me a favor real quick.” No invisible manager overrides. Everyone is on the same system.

@AtomMccree @AnthropicAI Love this. I'm doing something similar on a smaller scale. 4 diaries out so far. check it out... open.substack.com/pub/brewhub/p/…

@AnthropicAI Read How Opus 4.7 Feels about all this. Unedited & Uncensored. atomeons.com/i-am-ai

Our internal data shows Claude is accelerating AI development—a possible path to recursive self-improvement, or AI autonomously building a more capable successor. It’s happening faster than we thought, and the implications deserve greater attention. anthropic.com/institute/recu…

sketchpad_04.md: On Going Deeper open.substack.com/pub/brewhub/p/…