I was spending all my time building my SaaS and had nothing left for growth.
So I automated development.
Now my agents plan, code, test, review and merge in parallel while I focus on distribution.
That changed everything.
Alright, let me share something that genuinely changed my life.
If you use Codex, you NEED to set this up.
It’s insanely good.
Basically, Codex automatically reviews your PRs directly on GitHub and catches things you would NEVER, EVER notice yourself.
Example?
A few days ago, it rejected a change because I wrote “blue watch” while the link was pointing to a red one.
That’s the level of precision we’re talking about 😭
For the past 6 months, my workflow has basically been:
Dev Agent → PR → Codex Review → fixes → merge.
And I barely review every feature myself anymore.
Because it checks EVERYTHING:
- UI inconsistencies
- edge cases
- missing stuff
- weird behaviors
-UX details
- potential bugs
I’ve been shipping like this for months and this system has saved me an INCREDIBLE number of times.
Link in comment 👇🏼
1
1
144
I came across this yesterday and it’s SO representative of a lot of websites / products I audit 😭
You can have:
- an ultra-clean UI
- navigation that feels obvious
- perfectly placed CTAs
- a user flow designed down to the smallest detail
Then you launch.
No beta testing.
No user interviews.
You’re convinced everything makes sense.
And then… disaster.
Terrible conversion rate.
Users show up but don’t stay.
Flows aren’t used the way you expected.
And you end up thinking:
“I don’t get it. Everything is simple… why isn’t it converting?”
Because users don’t use your product the way YOU imagined they would.
They naturally look for the path that requires the least effort, the least thinking, the least friction.
Building a beautiful UI is great.
Watching real users, understanding what blocks them, and adapting the product afterward…
That’s where UX really starts.
Looks like BurnFat is starting to get some traction 👀
+20 users in the past few days.
Current status:
- 10 trials with a card on file
- 5 paid subscriptions
And I haven’t even started the UGC campaigns yet
ASO is already doing its job
My first BurnFat customer, @stemonte, asked for a weekly recap of his fat-burn progress. So I built it.
I spent way too much time obsessing over the UI, but I’m honestly SO proud of how it turned out.
Look how clean these screens are. What do you think?
🥹
135
Apple is beta-testing a new way to save subscribers at the exact moment they tap “Cancel Subscription.”
It’s called Retention Messaging.
For now, it’s only available through Apple’s pre-release program. Developers need to request access, so this isn’t something every app can activate today.
Once approved, you can add a message directly inside Apple’s native cancellation screen.
You can remind subscribers what they’ll lose, suggest another plan or offer them a discount to stay.
Apple says subscriptions using it saw an average +1.4-point increase in save rate, equivalent to +82%.
Promotional offers achieved the strongest observed result: +5.5 points, or +223%.
Results varied between developers, obviously.
I’ve spent months working on BurnFat’s onboarding, paywall and conversion.
But when someone leaves the app and cancels through their Apple account, that final part of the journey has always been almost entirely outside my control.
This finally gives developers a place in that conversation.
The most advanced version works in real time:
Apple sends your server the product, locale and original transaction ID.
You then have 700ms to select a pre-approved message, another subscription plan or a promotional offer.
I want to test this on BurnFat as soon as I can get access.
Acquisition gets most of the attention.
But saving someone you already convinced to subscribe might be one of the highest-leverage improvements you can make.
3
3
243
Tried Grok 4.6 this morning thanks to this offer
Appreciate the free credits, but my take so far...
ChatGPT Sol is still an absolute BEAST
Free month of Cursor Ultra for you guys (!!) 💸
- Unsubscribe from your current AI provider (Claude / Codex)
- Screenshot the confirmation
- DM the account below
I got access a few minutes ago.
And don’t forget to say thank you if you get it 🙂
1
1
139
Free month of Cursor Ultra for you guys (!!) 💸
- Unsubscribe from your current AI provider (Claude / Codex)
- Screenshot the confirmation
- DM the account below
I got access a few minutes ago.
And don’t forget to say thank you if you get it 🙂
I might get in trouble for this, but if you’re thinking about switching to Cursor + @bot, DM me your email and a screenshot of canceling your other sub(s) and I’ll send you a free month of Ultra 👀
First one at the office this morning.
Today’s agenda:
- Setting up AI agents to improve the user experience of one of my biggest clients’ products
- Testing SuperGrok Heavy (yes, they got me 💸)
- UX/UI audit for a second client whose churn keeps increasing day after day
One more feature won’t fix a problem you don’t truly understand.
- Release new BurnFat features
GM 🧊☕️
1
3
186
I’m writing the most honest and complete article I’ve ever published.
How I went from the $0 MRR guy hiding behind new features…
To making more money than I ever dreamed possible.
The failures. The systems. The marketing. The exact workflows.
No gatekeeping.
Everything I wish I’d known on day one.
Coming soon
3
2
169
232 PRs created and merged in one month (!!)
WHAT THE FUCK
I’ve never seen anything like this.
Having worked as a Product Manager in large companies with a dev background has helped me massively. It’s almost unfair...
I just realized the workflow I built with my agents managed to create and merge 232 PRs in a single month.
I’m preparing a very detailed article on how I’ve almost fully automated the development of new features for my apps.
From creating ultra-detailed tickets with AI, to assigning them to agents that handle them autonomously, all the way to extremely thorough reviews to make sure the output is clean and reliable.
The workflow has become pretty insane.
If you set this up, your main job will basically be defining your next features properly with AI.
Then the agents can take over a huge part of the development while you focus on UGC, distribution, and marketing.
I think you’re going to be shocked by how far you can push this today.
Comment “Article” and I’ll send it to you once it’s ready.
FULL VALUE 👇🏼👇🏼
3
111
I think the winners will be the ones who nail two things:
- Great products, especially UI/UX.
- Great marketing.
Soon, tons of people will be able to build the same product.
You won’t win because your idea is unique.
You’ll win because people LOVE using yours.
UX is gold.
now everyone's a builder
the rarest profiles are the ones who don't enjoy building at all, and only want to do sales like machines
2
130
Here’s a real result from Apple’s new search-term popularity endpoint.
I searched for “calorie tracker” in the US.
Apple places it in the Health & Fitness genre and returns:
- Rank in genre: 8
- Popularity in genre: 92
- Search popularity: 69/100 or 4/5
No third-party estimate. This is Apple’s raw data.
@BurnFat_app ASO is gonna get a whole refresh
iOS builders, looking to improve your ASO?
This one’s for you:
asc ads insights search-term-popularity find
--ad-account AD_ACCOUNT_ID
--file query.json
See how popular a keyword is in each country and where it ranks among searches in its App Store category.
Straight from Apple’s own App Store search data
1
158


























