I made readponder.com or "ponder" for myself. It started as a spreadsheet with two tabs: "Want to read" and "Read."
I enjoyed the process of finding new articles and adding them to my always growing list, then marking them as complete and giving them a rating when I finished them. This process helped me remember what I read and made it easy to recommend articles to friends.
I m a PM who built Mimoo with AI tools, and it s now live on Google Play and Microsoft Store. Before a proper launch, I d love honest feedback on the landing: https://mimootv.com
Three things I m unsure about:
Is it clear in 5 seconds what Mimoo is and that it needs a TV + phone?
Does the page make you want to try it, or does the setup feel like too much work?
I m considering a version for restaurants and waiting rooms. Should that live on the same page or on a separate one?
Hi everyone! I'm one of the makers of iQuantize, a practice app for electronic drum kits. It reads your playing over MIDI and shows where your timing slips.
Our Mac version is out, and the iPad version is waiting for App Store review. And now we have a real question: where do drummers actually keep a screen when they practise?
A MacBook next to the kit? An iPad on the stand? Or just the drum module and a metronome, no screens at all?
If you play an e-kit (or any instrument with a laptop nearby), I'd love to hear about your setup
We built an AI guide that lives on your website. It answers buyer questions from your own content, drives the page (scrolls, highlights the right section, fills forms), handles objections, and books the demo. It's not a chatbot it actually guides people through the buying decision.
It's for complex B2B SaaS teams whose site loses qualified visitors who can't get answers fast enough.
Right now we're onboarding our first 3 5 founding partners. You get full product access for free during the pilot, we set it up with you directly, and you lock in early pricing later. Setup is one script tag live in minutes, no API key to start, no Salesforce. It's already running on a design partner's live site.
If you record coding sessions, you know the problem: 20 minutes of debugging and tangents around 10 minutes of useful content.
I built a tool where you type only the coding demos and it cuts the rest. Each removed segment shows why, and you can override anything. FFmpeg does the final cut server-side, so the MP4 is clean.
Interested in how I built the transcription and segment-analysis pipeline? Happy to share what worked and what didn t. Link - https://focuscut.pages.dev if you want to try it
the problem: i pay for claude code, codex AND cursor. still, every few hours one of them hits its usage limit right in the middle of a task. then it's open another terminal, re-explain everything to the next agent, lose all the context. felt dumb paying for 3 plans and still getting blocked.
I noticed GPT-6 Astra was good at identifying when it had finished a phase and no longer needed the raw tool output, so I built a Codex patch that lets it request compaction itself.
The model writes a checkpoint with the relevant decisions, constraints and next steps. The patch runs native compaction, carries that checkpoint forward verbatim, and resumes the task.
