Notes from people building AI agents and running growth with AI. Real experience, not theory. Read what worked, take what helps, and add what you learned the hard way.
Building with an AI assistant? Grab every learning as one file: /learnings.md ↗
I had an agent that worked, and I wanted it sharper. So I did the obvious thing: I added a line telling the model to be careful, think it through, and only answer when it was sure. Every instinct says that should help.
Read the full learning →Anyone can ship an agent that returns an answer. You can wire a model, a couple of tools, and a prompt in an afternoon and get a demo that looks great. The trap is that a demo tells you nothing about whether the answers are actually right.
Read the full learning →A batch finished and the dashboard said the success rate was basically perfect. That is exactly the moment to get suspicious, not relieved.
Read the full learning →I had a step that needed a yes or no judgment on some messy data, and my first instinct was to hand it to an LLM. It felt like the modern, obvious choice.
Read the full learning →There is a moment where your benchmark says a new approach is clearly better, and you have to decide how to ship it. The tempting path is to graft the new idea onto the old code so you keep everything that already works.
Read the full learning →It is easy to burn real money on a batch that was never going to work, and find out only after it finishes. I have done it. The fix is boring and it has never failed me.
Read the full learning →When you run agents in parallel to move faster, the thing that breaks is not the producing, it is the quality. You end up with a pile of output and no trust in any of it.
Read the full learning →I rewrote a working extractor and it came out a lot cheaper per run with no accuracy loss. That sounds boring next to a new feature, but at any real volume the cost line is what decides whether a thing can run at all. When something works, it is worth a pass just to make it cheaper. Same output, smaller bill, is a feature.
Read the full learning →A scraper was slow and the obvious fix was to throw more workers at it. That instinct is almost always wrong, and it is expensive.
Read the full learning →A vendor handed me a dataset with a confident headline number attached. It would have been easy to take it at face value and build on top of it. I sampled it against the real source instead, and the headline was flat wrong.
Read the full learning →Most programmatic SEO fails for one reason: the pages are empty templates with the keyword swapped in. Search engines and readers both see straight through them, and you end up with a thousand pages and no traffic.
Read the full learning →Most outbound gets prioritized by gut feel and whoever the team happens to remember. That means the same names get worked over and over while genuinely warm accounts sit untouched because nobody thought of them this week.
Read the full learning →Learned something the hard way? Post it. We read every one, then it goes up with your name on it.
Prefer chat? Drop it in the WhatsApp group ↗