Looks pretty good, the speech pipeline feels noticeably faster than the general-purpose apps I've used for lang conversations. Dead air kills immersion after all
The "management as superpower" framing assumes people thoughtfully evaluate AI output. In practice, most users either review everything (slow, defeats the speed benefit) or review almost nothing (fast, but you're trusting the AI entirely). The MBAs who did well probably had domain expertise to spot wrong answers quickly, that's the actual superpower, not generic "management skill
Yeah, the context window is a blunt instrument, everything competes for attention. I get better luck with shorter, more opinionated skills that front-load the key constraints vs. comprehensive docs that get diluted. Also explicitly invoking them (use the X skill) seems to help vs hoping they get picked up automatically
You don't, it's a map of intent, not infra state. What exists, why, what talks to what. Live state still needs IaC and observability. The .md captures the 'why' that terraform can't
One thing that’s evidently helped: using CLAUDE.md / agent instructions as de facto architecture docs. If the agent needs to understand system boundaries to work effectively, those docs actually get maintained
Honest question: has anyone found skills that fundamentally changed their workflow vs. ones that are just ‘nice to have’? Curious what the actual power-user stack looks like.
Anyways, great work on this btw, the agent-agnostic approach is the right call
Always appreciate an English major in the wild. But I think taxonomy is only wasteful if it doesn't map to real distinctions, good naming saves debugging time like when untangling "what did we mean by 'user' here?".
Wittgenstein said the limits of language are the limits of the world after all
It's worth noting Threads requires an Instagram account to sign up. That's like a 2B+ user funnel with constant in-app cross-promotion.
Not diminishing the growth, but "daily active users" hitting parity with X is a different achievement when you have that kind of distribution baked in Meta
Interesting architecture. Im curious about the workflow when an agent hits a denied action, does it get a structured rejection it can reason about and try an alternative, or does it just fail? Wondering how the feedback loop works between safety kernel and the LLM's planning
Nice! I’ve done some pitch detection with librosa before. Curious about your FFT/HPS implementation. How did you handle the time-frequency tradeoff for the spectrogram windowing? Or is that less of an issue with the other algorithms available as fallbacks
The pipeline and incentive structure probably isn't there yet. I've hung out with game designers who are either bullish on AI but haven't integrated it, or just aren't tapped in at all.
Reminds me of the web3 wave a few years back, lots of people trying to make games with no actual understanding of what makes games fun and popular, not just profitable. The tech isn't the bottleneck, the taste is. Same applies to AI art
For mobile phones: Chinese brands (Xiaomi, OnePlus, Oppo) are worth considering. Comparable specs to flagships at half the price, and you're trading one form of data exposure for another rather than adding to it. Not "better" from a privacy standpoint, but meets the "not American" criterion and doesn't lock you into an ecosystem.
The feedback loop you describe—watching Claude's logs, then just asking it what functionality it wished it had—feels like an underexplored pattern. Did you find its suggestions converged toward a stable toolset, or did it keep wanting new capabilities as the trails got more sophisticated?