Launching something new always brings a weird wave of doubt right before it goes live. I started a habit a while ago of just dumping every worst case thought onto a blank sheet of paper like what if the server crashes or what if nobody even notices.
Once it s written down it somehow stops feeling like this massive invisible monster in my head. Most of the time the things I'm stressing about never even happen.
Do you have any little mental tricks or habits to calm your nerves right before a public launch?
I keep getting feature requests from people who want to use the product in a way I never intended. Not better versions of what it does completely different jobs. They want it to be more configurable, more customizable, more "just give me the building blocks and I'll figure it out."
Part of me wants to say yes to all of it. Flexibility widens the market, makes the product valuable to more people, and means I never have to turn a customer away. The more it can do, the harder it is to outgrow.
After OpenAI's DevDay, ChatGPT can now suggest an app in the middle of a chat, run it right there, and let people sign in with their ChatGPT account. 1B+ people open it every week, and they don't have to search an app store to find any of these apps.
I work in growth for an AI startup, so I keep asking myself where the next million users will find software. Will it be in a store, through Google, or inside a chat?
We keep a human in the loop for safety. It s the standard line during demos, and on paper, it makes total sense.
But in practice, human review quickly becomes the bottleneck. If someone has to check every step or approve every small update, the speed advantage vanishes. Worse, alert fatigue sets in people stop reading traces and just start blindly clicking "Approve."
When evaluating HITL setups now, my test is simple: Can someone make a confident decision in under 3 seconds?
If reviewing takes as long as doing the task manually, the agent isn't saving time it s just turning real work into an auditing job.
Working with busy people taught me a hard lesson nobody has the time to read your backstory. I used to spend twenty minutes crafting the "perfect" email flow only to get a single-word reply back. At first it felt kind of rude, but then I realized they were just respecting my time by keeping it brief. Now I try to put the main ask right at the very top so there's zero guesswork.
What's your personal rule for keeping messages short without sounding abrupt or rude?
Load testing always feels a bit like guessing games because synthetic traffic never really matches how actual humans behave on a live site. You can set up all the staging environments you want but real users will click things in an order you never expected and hit your endpoints all at once. I'm always curious how smaller teams handle this before pushing a major update.
Do you bother with heavy load testing tools or do you just patch things as they break on launch day?
I tried relying on the auto memory features built into most chat tools to remember my project preferences but half the time it saves random details I don't care about and forgets the stuff that actually matters. Then it starts bringing up outdated context from three weeks ago into a brand new conversation which is super annoying.
Do you turn those built in memory features off and start fresh every time or have you found a way to make them work reliably?
Before every launch I write one page about what I am afraid of.
Not a plan. Not a checklist. Just the honest list of everything I think could go wrong and everything I am secretly hoping does not happen. The thing I do not say out loud on launch day.
