kumailht.com
Helping founders design & build stuff:
http://kumailht.com
158 posts
605 followers
1,613 following
Regular Contributor
Active Commenter
comment in response to
post
yeah seems like it!
comment in response to
post
Looks sick!
comment in response to
post
Hmm, thanks for that explanation!
Excited to so some more coffee experimentation at home. I love going down this rabbit hole haha :)
comment in response to
post
I never understood the difference between a pourover and an aeropress/frenchpress?
Aren't they both doing immersion brews, would assume it's the same flavor profile?
comment in response to
post
What grinder do you use?
comment in response to
post
I’ve never made one. What device do you use?
I’ve been making Aeropress for years, but wanna try something new.
comment in response to
post
I think so!
comment in response to
post
Manual testing IMO suffices for most apps
Integration once it gets a bit complex
Unit tests for mission critical parts only
Typescript if you REALLY cannot afford to fail ever along with unit+integration+manual
But I think very few orgs qualify for the nuclear option
comment in response to
post
Love the design Andy!
Haven't seen stuff like this in a while, reminds me of hlvticons.ch
comment in response to
post
I have done it both ways and failed at both, but I still think the theory is valid!
comment in response to
post
Great point Jason!
comment in response to
post
🇺🇲 Google ---> 🇪🇺 Ecosia
🇺🇲 Chrome ---> 🇪🇺 Vivaldi
🇺🇲 Translate ---> 🇪🇺 DeepL
🇺🇲 Gmail ---> 🇪🇺 Mailfence
🇺🇲 Maps ---> 🇪🇺 Here WeGo
🇺🇲 ChatGPT ---> 🇪🇺 Le Chat
comment in response to
post
😂
comment in response to
post
It's got a cyberpunk space theme to it..
I LOVE how it turned out.
What do you think?
comment in response to
post
But it didn't.
The reddit user wanted a service for free, even though they clearly get value out of it.
comment in response to
post
Follow along to learn how I work on multiple high consequence builds myself and take them into production.
Will be sharing 2 more projects in the coming weeks.
comment in response to
post
The launch plan is:
1. Take to staging
2. Test internally for 1-2 weeks
3. Roll out just one product to production
4. Test in production for 2-3 weeks
5. Roll out the rest of the products
6. Continue to test/fix bugs/etc
The goal is to build confidence in the system over time
comment in response to
post
This project has been months in the making!
It's been so much work, but it's looking great now and was worth it. I think SoleMio customers will love it.
For builds this complex, you can't just throw it into production.
Too much is at risk. Millions in revenue, reputation, etc
comment in response to
post
The Production setup looks like this:
- Node
- Tailwind
- Railway
- Supabase
- Postmark
- ReCaptcha
- Stripe for Payments
- Custom Auth, Admin, Payments flow, Booking management & more
comment in response to
post
Hmm, yeah it's always a tough one!
I struggle with decisions like this too.
Do you think the multiple could be higher?
Do you have multiple offers?
comment in response to
post
That's a fair multiple offer at 3x ARR.
Are you going to go through with it?
comment in response to
post
And that's why I do all my client work myself. It keeps the quality up and keeps my business simple.
comment in response to
post
4. Clients hire me for what I can do. I can design world class software, keeping business context in mind, and also build and take that software into production.
It's what clients value about my work. And outsourcing that would defeat my core value proposition.
comment in response to
post
3. Hiring senior devs would work, but only at scale. And I have no interest in managing people and making more complexity for myself. I like simplicity.
Not to mention, senior devs are expensive, and I'd end up making near zero profit. So, pointless, unless you plan to scale.
comment in response to
post
2. AI is a GREAT junior engineer
If I were to hire someone affordable, I might as well ask AI to do it. It's faster, works 24/7 and tries their best at solving the problem. All qualities, a junior dev wouldn't have.
comment in response to
post
With Alex's permission, I'll be sharing how we design and build this.
Stay tuned for the next post!
comment in response to
post
This is what we came up with, it's a general direction of where we're headed.
The next step is to take this "low fidelity" mockup and turn it into production ready high fidelity designs that we can build against.