I hope I can get some advices here. I am developer, a solo developer. I make my project, snapencode.com. It is a self-host video platform. The idea is you buy license one time, for lifetime. No more monthly pay. You own it.
The building part, is fun for me. I love code. The logic.
What breaks when your AI agent needs a human answer?
I have been seeing the same pattern across coding agents and desktop agents: the agent can do more of the work now, but the flow still depends on small human approvals at the worst possible moments.
A shell command needs permission. A file change needs review. A calendar action feels too risky to run automatically. A tool call looks safe in one repo but dangerous in another. The agent pauses, the user is away, and the whole run loses momentum.
There were two clocks running on our launch and only one of them was under our control.
The product clock compresses. Working build, landing page, payments, guidelines. All of that is a checklist and you move it faster by putting more hours in. We had that side finished.
Had a thread rejected recently, still not fully sure why. Got me wondering about a few things and figured i'd just ask directly instead of guessing.
Does getting a thread rejected affect your account in any way, algorithm, visibility, standing, anything like that? or is it just a one-off, no consequence, try again kind of thing?
Also curious how the approval process actually works. is it a human reviewing each thread, some automated filter, a mix of both? trying to understand what actually triggers a rejection so i'm not accidentally repeating the same mistake without knowing what it was.
If anyone from the team or anyone who's been rejected before can share how this actually works, would genuinely appreciate it.
Someone asked here this week how to turn free users into paying ones. The answer I gave was half right, so here is the other half.
The easy half is that a free user has not decided the problem costs them anything yet. So you look for the ones with money already attached. A consultant, a tool they cancelled, someone's hours.
Hours are where it goes wrong. Every project I have been paid for was client work, priced before it existed. An hour saved there widened the margin, because the number on the contract did not move when the work got faster. That is why saving time reads as obviously valuable to me.
An agency billing hourly is the exact opposite. Every hour you save them is an hour they cannot invoice. Same job title, same daily pain, opposite reason to buy you.
I ve lost count of how many habit tracking apps I ve installed and eventually deleted over the last few months.
They usually start off feeling super motivating for the first fortnight. But as soon as life gets busy and I inevitably break a 10-day streak the whole experience turns into a guilt trip instead of a helpful tool. The constant daily reminders just end up feeling like spam notifications.
Lately, I ve been thinking simple physical journals might actually work better than phone alerts.
How do you keep track of your daily goals without relying on gamified apps?
I run more than one business, and for a long time I could not have told you, with a straight face, exactly which one was profitable in a given month. I had a general feeling, but no real number I trusted, just totals across bank accounts that never told the full story.
If you are running two or more businesses right now, do you actually know this month's numbers by business, or are you working off a general feeling too? What made you either build a real system for it, or decide it was not worth the effort yet?
Building something in the music space and constantly asking myself did I actually use this tool the way a real customer would? Or did I build for what I thought made sense? Curious what products you've bought recently that felt like the founder was a real user first, developer second. What tipped you off?
Text LLM serving is a banquet kitchen: you know the menu early, can batch requests, chunk prefill, and reuse common context through KV caches.
Streaming voice is a roadside dosa stall.
The user changes their order mid-sentence. VAD is separating speech from traffic. STT is revising words live. TTS needs to start speaking in ~50ms or the conversation feels broken.
