We've been stuck in merchant verification for almost a month.
The provider's site says a few days.
First the company took weeks to verify. Then the identity of the founder. Now they won't enable payment processing because they can't see the connection between our marketing site and our app, which lives on a subdomain of the same domain.
Same brand, same company, same documents they already have. One is the website, the other is the product. That's how software has worked for about fifteen years.
Has a press release ever actually brought you a customer?
For a recent launch I sent a free press release out through a handful of distribution sites. One of them got it onto AP News, a couple hundred local news sites and Google News within a few hours. No journalist wrote about it though, it was all automatic reprints.
What did come back was a bunch of sales pitches, an affiliate network asking to list the product, and a sign-up from a company that found us through ChatGPT. I'd done the same thing for my Mac apps before and got nothing like that, so my guess is people watch those feeds for B2B tools.
Have you tried one for a launch? Did anyone actually reach out because of it?
with AI agents helping us refactor modules and add endpoints rapidly, our team repository evolves much faster than before. the upside is quick execution, but internal setup guides and API docs quickly fall out of date.
team members frequently hit outdated configuration steps because the code moved forward while the readme files remained static.
My sessions no longer rely on each other for content.
More memory sounds good until you work from it every day. Then two questions present themselves. Who was it written for, the person or the next session? And is it a record, or something to act on?
I had chat memory carry the wrong language from one session into every new one. I could open each topic and rephrase it, and that did help. But it meant tracking two places, and I never knew which one a phrase came from.
So I turned chat memory off. What carries over is one record I keep: facts, decisions, notes, in plain files I can read. When something needs cleaning, I open the file and rephrase it, or keep the facts and start fresh. One source, so I always know where it came from.
I ran into a messy pipeline issue recently where an LLM was supposed to clean up incomplete user inputs before a database lookup.
The problem was that downstream functions started treating the model's cleaned-up version as the actual ground truth instead of just a search hint. It ended up matching fake data with real database entries because it was basically judging its own guesses instead of what the user originally gave us.
Now I keep raw user data completely locked away from any AI modifications during the final check.
How do you keep your AI outputs isolated so they don't corrupt the actual logic?
Our free invoice generator takes a JSON post with no account behind it and hands back a PDF. So it has limits on everything.
At most a hundred line items. At most twenty taxes on any one item. A length cap on every text field, a byte cap on the logo, and checks that run before anything expensive is decoded. Each limit was chosen by asking how large one field can reasonably be, and each one was sensible on its own.
Then we measured a request that respected every single one of them: a hundred items, each carrying twenty taxes. It answered 200, and it held a worker for about 62 seconds. The same hundred items with one tax each took about 2.4.
Nothing was broken. The limits multiplied. A hundred items times twenty taxes is two thousand line and tax pairs, and each one is printed in its line and, when the names differ, again as its own row in the totals. The item limit counts items. The per item tax limit counts what sits inside one item. Nothing counted the document.
