General
p/generalShare and discuss tech, products, business, startups, or product recommendations
trending
•

3mo ago

Your AI subagent's report is all you'll ever see. Make it hand in receipts

If you use AI agents that spawn helper agents (audits, research, big refactors), here's the mechanical fact that changed our setup: the parent never sees the work. The helper runs in its own context; when it finishes, one final message comes back. The forty tool calls behind it the searches that found nothing, the test that failed twice before passing, the edge case quietly ruled out of scope stay behind. What you receive is a summary, compressed by the thing being evaluated.

That's not a bug. Isolation is why delegation is affordable at all. But it means every "Done, all tests pass" is testimony, and nothing in the loop charges the helper for optimism. A caveat is a paragraph; "done" is one word. Guess which survives compression.

Ours got caught the memorable way: an audit subagent confidently reported that a sourced claim had no source behind it. The source existed. We found the exact sentence in the original docs and ended the argument in minutes.

What we require now, roughly in order of payoff: evidence formats written into the dispatch prompt (every finding arrives with a verbatim quote and a file path), artifacts over adjectives (test output and re-runnable commands instead of "thoroughly checked"), and spot-checks scaled to blast radius anything we'll act on gets one finding re-derived end to end. The re-check is deliberately cheap: when a claim comes holding a quote, verifying it is a text search.

•

3mo ago

Which SaaS metric did you measure wrong without noticing?

Calculating SaaS metrics is my weekend hobby, same as our family budget and the mortgage plan, which is a strange thing to admit. My conclusion so far: almost every number I report has a definition I can move, and the flattering one always survives.
Nobody lies. The definition just drifts.
MRR: annual contracts divided by twelve, correct, and setup fees in the same total, not correct. Nobody decided that, both arrive as money in the same month.
LTV over CAC is the one that cost me something. Both inputs were already soft and I was treating the result as a decision. A wrong ratio does not look wrong, it is a normal number in a normal range.
So which one did you measure wrong? Not on purpose for a deck. The one where you found out later the definition had moved.

•

3mo ago

I launched twice this year and got 3 upvotes total. Here's what I actually got wrong.

Two launches in 2026:

HalalNear.me (Feb): a halal restaurant directory, 1,700+ verified listings across 8 cities. 3 upvotes. Zero comments.

•

3mo ago

How do you launch multiple products on Product Hunt?

Hey everyone!

I ve already launched my website on Product Hunt, and now I m planning to launch individual products under it.

What s the best way to do this? Should each product have its own Product Hunt launch, or is there a way to launch products under an existing company?

Would love to hear from people who have done this before. Thanks!

•

1yr ago

Text to voice (reader) for threads + ugc

While on a walk yesterday, I wanted to learn about the Spain/Portugal blackout. Hacker News had a great thread on the blackout. I used ElevenReader (amazing product by @ElevenLabs) to listen to the thread. Overall, it worked great, but:

  • The reader treated new comments as just new paragraphs. So I had no idea when the reader transitioned between comments/responses. Usernames of commenters were not mentioned.

  • Urls were read out fully. This is brutal.

  • It would be awesome to skip to next comment with a voice command.

Any product recommendations for this? Or maybe @ElevenLabs is already thinking about this?

•

3mo ago

Does anyone have a strong opinion on where the IRL economy is going?

I'm guessing many of you have seen it already.

Pretty much every consumer company that markets itself as some version of "helping you spend time with friends in the real world" is becoming an event discovery app. I thought event discovery was a tarpit idea and we can't just have 13 different places to discover things to do.

•

1yr ago

How do you know if your cold emails are working before users reply?

I'm building a tool that uses AI to help craft cold emails that actually get responses. But before I go too far, I want to dig deeper into the real frustrations behind cold outreach.

Most founders and marketers I talk to say something like:

We send cold emails, but we re not sure if they re hitting the right tone, targeting the right pain point, or just going straight to spam.

Right now, I m researching the before you send side of cold outreach that moment where you're staring at your draft, wondering if it s persuasive, personal, or just meh.

•

1yr ago

Where startups are spending on AI according to a16z and Mercury

Full report from @olivia_moore @mha1 & @seema_amble :

  • Horizontal apps have a slight lead over vertical (60% of the list). This includes general assistants (ex. @Perplexity) and SIX different meeting support tools (ex. Fyxer AI). But, it also encompasses creative tools and vibe coding tools that are used in roles across orgs.

  • Vertical apps can augment human labor...or replace it. We're mostly seeing the former - but five companies on the list allow customers to "hire AI" (ex. Crosby Legal, @Cognition IP , @11X). Labor augmenters mostly assist with customer service, sales, and recruiting.

  • Vibe coding has landed in enterprises. It's not just a prosumer trend! Number three on the list, below @OpenAI and Anthropic? Replit. Other listmakers in the category include @Lovable and @Emergent 2.0, while @Cursor made the ranks for more technical users.

  • Products are making the consumer -> enterprise jump. 12 cos also appeared in our most recent Consumer AI Top 100 - almost all of which started out B2C and have migrated B2B over time. In fact, 70% of listmakers are available for individual use (no enterprise license needed)!

•

3mo ago

Why does using AI-writing get judged differently than using any tool for literally anything else?

Nobody questions someone's competence for using a calculator, a spellchecker, a design template, or an IDE autocomplete. Using a tool to help execute an idea is normal everywhere else.

But the second writing involves AI even when the person had the actual idea and used AI just to help draft or refine it there's an instant judgment that doesn't apply anywhere else.

•

3mo ago

How do you vet overseas apparel/clothing manufacturers before placing a bulk order?

I've been on the supplier side of apparel sourcing (based in Hangzhou, working with winter clothing manufacturers), and I'm curious how brands on the buying side handle vetting before committing to a factory.

A few things I'd love to hear from this community:

- How do you evaluate fabric/material quality before seeing a physical sample?

First
Previous
•••
500501502
•••
Next