Test

A lot of folks (including myself) are creating iOS apps these days. So... I created a FREE app store screenshot editor: appshoteditor.com No sign up needed. Just create your multi-screen horizontal screenshots -> Export -> Done Right now it only has the iPhone 16 mock up, but adding more today

The coder LLM isn't doing a good job. Time to pull out the big guns.

“This model is so goooood…” AI models are probabilistic. It’s like flipping a coin and getting 10 heads in a row. Eventually you’ll hit tails — and suddenly people think the coin is broken.

“This model is so goooood…” AI models are probabilistic. It’s like flipping a coin and getting 10 heads in a row. Eventually you’ll hit tails — and suddenly people think the coin is broken.

My laptop battery doesn’t last long now. Not really battery health issue. Doing work in parallel really drains battery fast.

Good Claude. Good. cc @jeremysmith.co

They knew if they let it slide, they'd lose the UI -- which is the key. Models will be commoditized as everybody's models improve over time. Important UIs of the past: - Desktop OS - Mobile phone platform - Browsers - Social network du jour

Want to set a deadline that you'll definitely meet? Give yourself a deadline. I just posted to my customers on my product's private Discord that v2.0 will be out in 7-10 days. Now, I'll move heaven and earth to get this shipped.

People are still saying "I never heard of any successful vibe coded app." Here's my take on this:

I'm selling some domains: bulletpaste .com growemail .com guideful .io tenantssl .com workorder .io widgetful .com If anyone is interested, send md a DM.

Does anyone play Big 2, Tiến Lên, or Dou Dizhu? Built some card games for fun. Link: bs.fyi If you do play, let me know if you notice any bugs!

An honest footer... LOL

They say products need to be launched again, and again, and again. I built Broadcast back in October 2024 (hand coded). It's an email marketing platform, and it's matured quite a bit.

It is the year 2027. There are now 3.4 trillion apps. On average, there are about 400 apps for each person on the planet. 99.99999999706% of apps have never been downloaded even once. Estimated iOS app store review wait is about 42 years.

When was the last time you subscribed to a SAAS? Or did you just head into Claude Code and built yourself a custom solution instead?

If people don't hold unnecessarily strong opinions about your open source project (without ever contributing in any way), is your open source project really *open source*?

Heart breaking to see this. Tailwind CSS is such a useful tool, and used by many. Its business model no longer works in this post-AI world, and their business isn't the only one. I've talked with several founders about how AI has killed their business. It isn't discussed openly, but it's real.

Accidentally logged out of bsky. 2FA email ended up in spam. I really hate spam filtering. Yeah, I know what it's supposed to do, but it's like some other person telling me what emails I should get.

One day, nobody will know how to fix broken code. Just like most people don't know how to fix their cars. Or do plumbing trade etc. Then, whatever-bros at that time will say, "Yeah, the best jobs are those blue collar JavaScript programmers. Let's all go and be JS plumbers."

Deep extinction thoughts from my kid: "I think I get it now. Dinosaurs were real, but are not real anymore."

I would recommend sendbroadcast.net built by @geetfun.com with Ruby on Rails

Released v0.17.0 for sendbroadcast.net Get 25% off licensing with code NEWYEAR25 Things added/fixed this week: • Added ability to import CSV files into subscribers • Fixed bug where AWS SES webhooks were not processing correctly • Updated server dependency • Updated internal dependencies

Added some charts to sendbroadcast.net Will be adding more visuals like this in other parts of the app

Tell me you were unproductive yesterday without telling me. I'll go first. I didn't even know GitHub was down.

It's likely 100% true that the thing you're building already exists

Well, it's time for me to start a new project. Last night, while working on my _other_ project, I realized I needed a tool for it to test some things out. Fingers crossed I can get it done in 3-4 days.

Just sent my 13th weekly product update to my customers for sendbroadcast.net. It's become a habit now that every Sunday night, I look at my changelog and summarize what I've updated/improved/fixed and fire off an update to my customers.

Released v0.14.1 for sendbroadcast.net • Sequences can now use conditions to direct the flow of DRIP/autoresponders

Been reading the book, The Psychology of Money. An idea in there is that it's oftentimes hard to know how much of a result is attributable to luck or skill. As a result, this makes it hard to extrapolate the lessons from someone else's outcome to apply to your own situation.

Working on sendbroadcast.net, I am constantly reminded how glad I am to have spent the time writing unit tests. At the start of any side project, it's so easy to just code. But as complexity increases, tests are what gives me the confidence to ship.

Digital Ocean calls servers "droplets" Linode calls servers "nodes" Hetzner calls servers... "servers" Just curious how people feel about companies using their own language/vocab?

My December 2024 report for sendbroadcast.net Yep, I'm doing this 7 days into 2025 :-) 💰 Made 3 sales (m/m 0%) 💻 70 commits into the project (prev 96) 🌟 Domain rating 10 (prev 11) 🚀 Added 10 features incl: • ESP integrations • Improved preview rendering • Resumption of interrupted broadcasts

Just updated the footer to © 2025.

First update for sendbroadcast.net in 2025... Multiple internal dependencies updated!

Only 363 days left this year to make stuff happen

I spent the day restringing 4 tennis racquets 🎾

Happy New Year everyone!

Got some work done tonight for sendbroadcast.net! Now possible to view recipient activity on email broadcast campaigns at the subscriber level...

One more release before I take a break for a few days... :-)

Just sent the 11th product update for @sendbroadcast.com to customers... using sendbroadcast.net of course. One more product update to give for next week, so looking forward to knocking some of those todo's on my list. Let's go Monday!

Released v0.9.1 of sendbroadcast.net • Added Mailgun webhook integration for handling email opens and clicks and other metrics • Fixed bug in Postmark webhook integration tracking • Upgraded to Rails 8.0.1

Who has created an internal tool and turned it into a product?

I’m still alive folks! Just in build mode at the moment. 🔨

Almost 1 release per day this week. I think I'm going to take Sunday easy.

Sometimes customers will ask questions like “How is your product different from __?” And sometimes there’s really no straightforward answer.

Released v0.6.2-0.6.3 with some small bug fixes and optimizations: • Fixed bug where adding tags to subscriber did not trigger adding to sequence • Fixed subscriber count in sequences • Sequence restarts set at 30 seconds instead of 10 minutes Full changelog at sendbroadcast.net/changelog

Why start side projects even when you have a main one? The reason is that side projects give you a chance to try out new tech. It’s an excuse to try and test something new. Sometimes techniques or tech from a side project is so useful, you can bring that knowledge back into your main project.

Building a business that can get acquired should be a goal. Even if you don’t want to get acquired. Think about it: you want your business to become valuable enough that someone wants to buy it. To have processes that it can run without you. Those should be the goal anyway, right?

So Monday to Tuesday, I was fixing a bug. Specifically it was to handle auto saving in a text editor. Initial implementation took me something like 45 mins. But edge cases cropped up. Technically things weren’t hard, but the logic took 2 days to think through. And that’s dev work for ya, folks.