Been offering a free plan for two years now, and i still can't decide if it's helping or just wasting time.
It brings in steady word of mouth traffic but free users generate most of our support tickets without ever converting. Paying for servers for people who will never upgrade is starting to feel like a bad trade.
Ember tracks five categories (sleep, fitness, discipline, focus, spending). My code picks which one gets asked next. My code compares your new answer to your last one and works out how the grade changed. My code decides when an achievement unlocks. The model gets those results and writes the roast around them, and it s only allowed to quote numbers that actually changed. For example, if your sleep grade went from C to B, that s the fact it gets. It doesn t get to guess.
I m not claiming this makes it perfect. But when a roast looks wrong, I know where to look: is the data the model received wrong (my bug), or did it write badly from correct data?
One related bug I hit: achievements kept un-unlocking themselves because three parts of the code wrote to the same row without checking each other. The fix was a compare-and-swap write, so the write succeeds only if the row is still what it was expected to be.
Where do you draw the line between what the model decides and what plain code decides?
We often cover software topics on ProductHunt, but sometimes we overlook the fact that tech can be felt most in the physical world (hardware).
As of mid-2024, there have been approximately 3,979 reported incidents involving autonomous vehicles (AVs) in the United States since reporting began in 2019. Source
I built my app in months. Marketing has been the hard part. There's no "vibe marketing" yet, but a few tools made specific jobs much easier. Here's what I used, task by task:
1. Launch video ElevenLabs
What I used: its image, video, voice, and music tools. I made one reference image of my character and generated every scene from it, so his face stayed the same. The voices use Eleven v3 dialogue with tags like [tired] and [pause].
Tip: avoid the AI agent editor. It used about 3x the credits I expected. Generating each piece yourself is cheaper.
It's so easy to get high on the dopamine of seeing signup numbers tick up every Monday morning, even when most of those users never log back in after day three. We hit a wall when we realized our massive growth spike was just filling a bucket with a hole in the bottom. Getting the team to stop caring about vanity metrics and actually fix the onboarding friction took a real mindset shift. If you've been through this, what metrics finally forced your team to wake up and focus on actual usage instead of just new signups?
Most AI features feel like gimmicks because they're a button looking for a reason to exist. Someone in a meeting said "we need AI" and three weeks later (or maybe hours now with Claude Code) there's a sparkle icon in the corner that summarizes things nobody asked to have summarized.
The test I use: if you stripped the word "AI" off it entirely, would people still want it? If the answer is no, it's a gimmick.
I've spent the last 1.5 years deploying AI voice and chat agents for banks, insurers, car dealers and online stores. I'll be here all day answering questions.
Ask me about anything: getting through enterprise security reviews, what breaks when an agent writes to a CRM, how we handle human handoffs, pricing, or what goes wrong in a first pilot.
No question is too basic. What would you like to know?
