jhiggins777

These aren't good. They have the same llm-speak that has pithy short sayings but are strung together not in a way that unfolds naturally and clearly but only as generally conceptually related.

Good explainer videos need a human teacher to craft the narrative and exposition.

I've kept a running list of what to do with extra tokens. Would love other ideas people have.

Nutshell subscribes you to RSS feeds and emails you a daily digest. Each morning it fetches new articles, runs them through an LLM, and sends you a summary to your inbox. The main configurable piece is the summary format. You can set writing style (bullet points, narrative, technical, casual, or a free-form custom system prompt) and reading length (short ~1-2 min, medium ~5-10 min, long ~10-20 min). Changes take effect on the next send, no re-generating history.

This is little croppy app.

The problem I was having was that when I was passing UI screens to coding agents, I know I should break down the screen into all the components and pass 1 component per turn in order to get the best component generation.

But this was annoying because you just have to do all the cropping in figma/photoshop. So I made this little utility app.

- Upload UI (Mobile or Desktop) - Passes it to AI to analyze into components - You fine tune cropping (this happens client-side, super fast) - Export images

Free - 1 Page $5 - 100 Pages

Let me know if you want any features.

...ai

Have you used it? I use it for both hands free and read later. When I'm on a webpage I just use the safari share sheet to send it to ElevenLabs Reader and then just listen whenever I have time.

Thought the same thing.

I, however, have not written any containerization code.