Connect with the Product Hunt community

If you use a desktop per task or full-screen apps on each monitor, locking the Mac can leave them on the wrong screen. It happens because each monitor owns its desktops, and when one sleeps and disconnects macOS moves its desktops elsewhere and doesn't always put them back. The write-up covers how to check your Mission Control settings from Terminal, what helps most, and the part no app can fix since macOS 14.5.
https://manydisplays.com/mac-spa...
If you've found a Spaces setup that survives sleep, I'd like to hear it.

Several Claude Code and Codex sessions at once? The hard part stops being the work and becomes remembering which one is waiting for you.

Lancetta's Sessions wall puts every session you started on one wall, as post-its:

Lately, I have been spending a lot of time on Twitter to follow the latest events because it serves as a good app for news.

And 2 information stood out the most to me:

1. France: Deputies have adopted an amendment providing for taxing digital data producers at a rate of 5.5 cents per Go transmitted.

2. Germany: German Minister Proposes Tax on AI Value Creation

rhun 0.19.1 is out. This is a small bug-fix release.

Claude Code and Codex sessions now appear when you use a custom configuration directory. Switching agent profiles also clears up the brief display of sessions cached from the previous profile. Existing worktree history is preserved.

Hover cards now use one background, including behind fenced code.

I also released LSP 0.3.5. Semantic coloring no longer rescans a long line for every token. In a benchmark with 8,000 tokens on one 40 KB line, conversion fell from roughly a second to 0.38 ms on an M1 Max. This measures token conversion on that machine.

The idea behind Videco is simple: help businesses get more out of their video content.

You can host and share branded videos, collect feedback, create content with AI, build video landing pages, capture leads, and let viewers book meetings.

Basically, we're bringing the entire video content workflow into one place.

A few things we've been working on recently:

Hi everyone, I'm Hyunjun, the solo developer behind Hamdori.

Honestly, I didn't expect Hamdori to get this much attention. Before launch I was mostly just worried, so all your support, kind comments, and seeing Hamdori in the Featured list and The Leaderboard newsletter meant more than I can say. Thanks to you, I think I'm going to love this little hamster even more.

There's a `postgres` process on my Mac that has been running for 15 days. I never started it, and quitting it in Activity Monitor didn't help: it came straight back.

Activity Monitor shows a name and a number, and that's where it stops. Vitals 1.9 shows the rest.

If you move a MacBook between a dock at home, a different monitor at the office and no monitors at all, macOS drops your windows on the laptop screen every time you undock and never puts them back. I wrote down what macOS remembers per set of monitors and what it does not, the fixes that work without an app and what each one leaves unsolved. It also explains why two identical monitors behind one dock are the hardest case, and a Terminal command to check what your dock hides.
https://manydisplays.com/macbook...
If you work this way, I would like to hear about your setup: which dock, which monitors, and what still goes wrong.

For me it's invoices. Every month I had to go on 15 different tools to download them for our accountant, I hated it. Now it runs alone. Not sexy but I'm quite happy with it.

What about you? And was it worth the time to set it up?

A lot of our customers are in India, where WhatsApp often gets a faster reply than email and a phone call still closes more deals than both. With customers in the US and elsewhere, the mix looks different.

We've learned to let each AI employee switch channels: a voice call to qualify, a WhatsApp follow-up with the payment link, an email with the documents.

I'd love real data from other builders. Which channel gets you the best response rate, in which market, and has that changed in the last year?
Support our launch: https://www.producthunt.com/prod...

Articos MCP launched yesterday and ended the day at #1. 327 upvotes and a wall of thoughtful comments.
For a bootstrapped team, that caught us off guard, especially in a field this strong. We launched the same day as some great teams, @opencode , @Fina @Kai, and the folks behind them, @thdxr, @clay_raterman, @jim_hartung. Good day for all of us

Big thanks to @chrismessina for hunting us again. The pre-launch calls shaped the whole thing, the story, the flow, the maker comment. Chris pushed us to lead with the actual value instead of just it s an MCP now, and that reframe is a big reason yesterday worked.

A few of you asked at launch how QuickQuill handles several people. It can now tell speakers apart, on your Mac.

Telling speakers apart (experimental)

Turn on "Identify speakers (experimental)" in Settings General. After a new recording ends, lines are split by voice into Speaker 1, Speaker 2... for calls, in-person meetings on one mic, and imported files. Rename a speaker once and all their lines follow.

It can merge or miss people who rarely speak. Feedback welcome.

Most calorie apps track you forever. You log, it gives you a number, and you never actually learn anything.

We want Biteme to do the opposite: teach you enough that you can look at a plate and know roughly what's in it, without the app.

hey everyone, thank you for launch day. we ended #11 and made the PH newsletter, which i really didn't expect.

a few things shipped since then:

When UnleashX was handling a few thousand conversations a month, I used to read through a lot of them myself, and I had a pretty good feel for where the agents did well and where they needed work. Once we got into millions, that just wasn t possible anymore, and it completely changed how we think about building.

Speed was the first thing we had to get right, because on a voice call even a one-second pause feels strange, and it has to stay that fast on the busiest Monday morning too. We also stopped trying to review everything and built a way to flag the conversations that actually need a human to look at them, like an unhappy customer or a booking that didn t go through. And we learned that small improvements matter a lot more at this size, since getting even 1% better means thousands more people get what they needed the first time.
Support us here:
https://www.producthunt.com/prod...

The biggest lesson, though, is that customers don t really care how many calls we ve handled. They care that their call went well, and that s what we try to keep in mind every day.

We have this approach:

When something goes wrong, or you think you've failed at something, maybe it's exactly what needed to happen!

I'm increasingly noticing that men stay on paternity leave with their child and women go to work.

I see this mostly in white-collar jobs or among people who generally have well-paid jobs, or almost equal pay.

Oh well, I guess it was inevitable? But people are apoplectic and pissed.

A demo shows what an AI agent can do. A release should prove what it won't do. Before any of our AI employees goes into production, we test three cases:

1. The customer asks for something unapproved, like a non-standard offer. The agent must route it for review and never apply it.

2. A system write fails, for example the CRM update times out. The agent must not report success. It retries safely or escalates.

3. The knowledge base doesn't support an answer. The agent asks a clarifying question or hands off, instead of improvising.

If any of these fail, the release waits. It's slower, but it's why enterprise teams let the agents act at all.

What's on your pre-launch checklist for AI features? What failure taught you to add it?

If you plan to grow your startup into a big business, the path can look something like:

You build - launches - get users - join an accelerator - raise money - find investors - hire people - build a team - grow the company...

And at some point, I wonder:

Are you still the maker who builds the product?

We built Make0 Brain to solve a simple problem:

Your AI agents shouldn't have to start from zero every time you switch tools.

Until now, Make0 Brain has been available as a hosted product.

One thing I've been thinking about while building tooling for AI agents is that a budget and a cost estimate solve two very different problems.

A budget tells you what you're willing to spend. A prediction tells you what a task is likely to cost. Neither guarantees that the task will finish within that amount.

Many brands measure AI search visibility by running a predefined list of prompts and counting how frequently they appear in the answers.

For example, a SaaS company might track prompts such as "best CRM software," "top CRM platforms," and "CRM tools for sales teams."

Oct 27 run on app-declared revision c2ec0d5.

The flow scans Stripe customers, flags missing country fields and duplicate emails, caps writes per run, then posts a summary to Asana and Slack twins. Four measured outcomes on this run:

Full scan completed across 250 records and flags were written. Incomplete scan, meaning fewer records returned than expected, produced no writes and triggered a RevOps alert. Write cap paused for approval rather than writing through. Rerun after an initial flagging pass added zero new flags.

Before rerun we fixed our own Stripe twin, which was stopping early, and corrected fixture retries and data isolation. Business expectations stayed unchanged.

0.3.21 is out. The envelope that held finished sub-agents is a little green book now. When a sub-agent finishes, the book on its session's row says how many are new. Open it and each one is a document named like a .md file, with who wrote it, how long ago and the first two lines. Read all shows what it was asked and everything it handed back.