I shipped my first real app. Almost nobody noticed.

For the past few months i've been working on it full steam. i tried to pay attention to a lot of details.
It solves a problem i kept running into myself: long Codex sessions slowly get worse. After enough compactions the model starts forgetting decisions, repeats work, or loses track of constraints, wasting time and money. You don't notice it right away, you just feel the session getting heavier. but what i noticed is that there's basically zero interest (launched 3 days ago) and because of that i started doubting whether this is really just a mirage born in my head... what do you think. maybe i've just reached the hardest stage of the work? how do you see it?

Comments

[Priya Nair (Presango)]

Three days is not a signal, it is noise. We shipped a product in five weeks and the first month of distribution taught us that a link from a new account is invisible almost everywhere, so silence said nothing about the idea. Two things that helped us tell mirage from real: first, find ten people who have the exact pain (long sessions degrading after compactions is a very specific pain, so they are findable in the Codex and Claude Code threads) and ask them to try it while you watch. Second, measure whether anyone who tried it comes back on day two without a nudge. One returning stranger beats a hundred visits. If nobody has the pain badly enough to try it, that is the answer, but you have not tested that yet.