Alex Lau
@mralexlau.bsky.social
Dad. Author of the Amazon #1 Best Seller "Keep Calm and Code On." Software developer of over 15 years. Meme curator. Slay the Spire fanboy.
The bad news? There are no shortcuts.
The good news?
You don't need to be a prodigy or genius to succeed.
In software, what separates teams and developers who deliver from those who struggle isn't magic - it's diligence, focus, and consistency.
The good news?
You don't need to be a prodigy or genius to succeed.
In software, what separates teams and developers who deliver from those who struggle isn't magic - it's diligence, focus, and consistency.
December 18, 2025 at 3:07 PM
The bad news? There are no shortcuts.
The good news?
You don't need to be a prodigy or genius to succeed.
In software, what separates teams and developers who deliver from those who struggle isn't magic - it's diligence, focus, and consistency.
The good news?
You don't need to be a prodigy or genius to succeed.
In software, what separates teams and developers who deliver from those who struggle isn't magic - it's diligence, focus, and consistency.
Developers have their own math... and it doesn't always make sense to anyone else.
- 10 minutes of coding + 2 hours of debugging → still only "a quick fix"
- If it's open-source → it's free, even if it takes me 3 days to configure.
- If it works in dark mode → it's fully accessible.
- 10 minutes of coding + 2 hours of debugging → still only "a quick fix"
- If it's open-source → it's free, even if it takes me 3 days to configure.
- If it works in dark mode → it's fully accessible.
December 11, 2025 at 3:03 PM
Developers have their own math... and it doesn't always make sense to anyone else.
- 10 minutes of coding + 2 hours of debugging → still only "a quick fix"
- If it's open-source → it's free, even if it takes me 3 days to configure.
- If it works in dark mode → it's fully accessible.
- 10 minutes of coding + 2 hours of debugging → still only "a quick fix"
- If it's open-source → it's free, even if it takes me 3 days to configure.
- If it works in dark mode → it's fully accessible.
What are your thoughts on footcannons and "sharp knives" in software?
I've gone back and forth over the years, but here's my take:
Some tools give you a lot of power.
But they also give you a lot of ways to hurt yourself.
I've gone back and forth over the years, but here's my take:
Some tools give you a lot of power.
But they also give you a lot of ways to hurt yourself.
December 4, 2025 at 3:06 PM
What are your thoughts on footcannons and "sharp knives" in software?
I've gone back and forth over the years, but here's my take:
Some tools give you a lot of power.
But they also give you a lot of ways to hurt yourself.
I've gone back and forth over the years, but here's my take:
Some tools give you a lot of power.
But they also give you a lot of ways to hurt yourself.
Some bugs feel supernatural.
You refresh the page for the 11th time, it's broken.
You refresh it once more... and suddenly everything works?!
Now the bug is gone, the logs look clean, and you're left questioning reality.
But here's the thing I've learned over the years:
You refresh the page for the 11th time, it's broken.
You refresh it once more... and suddenly everything works?!
Now the bug is gone, the logs look clean, and you're left questioning reality.
But here's the thing I've learned over the years:
December 2, 2025 at 3:01 PM
Some bugs feel supernatural.
You refresh the page for the 11th time, it's broken.
You refresh it once more... and suddenly everything works?!
Now the bug is gone, the logs look clean, and you're left questioning reality.
But here's the thing I've learned over the years:
You refresh the page for the 11th time, it's broken.
You refresh it once more... and suddenly everything works?!
Now the bug is gone, the logs look clean, and you're left questioning reality.
But here's the thing I've learned over the years:
This year I want to do TWO things special for Cyber Monday (as a continuation of Black Friday):
1️⃣ My ebook Keep Calm and Code On is on sale for just $1.99 - an incredibly low price for lessons I spent 15+ years learning the hard way (link in comments).
1️⃣ My ebook Keep Calm and Code On is on sale for just $1.99 - an incredibly low price for lessons I spent 15+ years learning the hard way (link in comments).
December 1, 2025 at 3:00 PM
This year I want to do TWO things special for Cyber Monday (as a continuation of Black Friday):
1️⃣ My ebook Keep Calm and Code On is on sale for just $1.99 - an incredibly low price for lessons I spent 15+ years learning the hard way (link in comments).
1️⃣ My ebook Keep Calm and Code On is on sale for just $1.99 - an incredibly low price for lessons I spent 15+ years learning the hard way (link in comments).
This year I want to do TWO things special for Black Friday:
1️⃣ My ebook Keep Calm and Code On is on sale for just $1.99 - an incredibly low price for lessons I spent 15+ years learning the hard way (link in comments).
1️⃣ My ebook Keep Calm and Code On is on sale for just $1.99 - an incredibly low price for lessons I spent 15+ years learning the hard way (link in comments).
November 28, 2025 at 3:01 PM
This year I want to do TWO things special for Black Friday:
1️⃣ My ebook Keep Calm and Code On is on sale for just $1.99 - an incredibly low price for lessons I spent 15+ years learning the hard way (link in comments).
1️⃣ My ebook Keep Calm and Code On is on sale for just $1.99 - an incredibly low price for lessons I spent 15+ years learning the hard way (link in comments).
There's always that moment when you find a bug and think:
Do I fix the root cause... or just opt for a quick fix?
One path takes longer but cleans things up.
The other gets you unblocked fast but adds a little more weight to the codebase.
Do I fix the root cause... or just opt for a quick fix?
One path takes longer but cleans things up.
The other gets you unblocked fast but adds a little more weight to the codebase.
November 25, 2025 at 3:00 PM
There's always that moment when you find a bug and think:
Do I fix the root cause... or just opt for a quick fix?
One path takes longer but cleans things up.
The other gets you unblocked fast but adds a little more weight to the codebase.
Do I fix the root cause... or just opt for a quick fix?
One path takes longer but cleans things up.
The other gets you unblocked fast but adds a little more weight to the codebase.
Why are software estimates almost always too short?
Because we tend to estimate the happy path -
The version where everything works, nothing breaks, and no one gets interrupted.
We forget:
- That test that fails for no reason
- The merge conflict that eats an hour
Because we tend to estimate the happy path -
The version where everything works, nothing breaks, and no one gets interrupted.
We forget:
- That test that fails for no reason
- The merge conflict that eats an hour
November 13, 2025 at 3:02 PM
Why are software estimates almost always too short?
Because we tend to estimate the happy path -
The version where everything works, nothing breaks, and no one gets interrupted.
We forget:
- That test that fails for no reason
- The merge conflict that eats an hour
Because we tend to estimate the happy path -
The version where everything works, nothing breaks, and no one gets interrupted.
We forget:
- That test that fails for no reason
- The merge conflict that eats an hour
There's an idea in software that's stuck with me over the years:
⬇️ It's called the "Prime Directive" - and it's not about Star Trek ⬇️
The prime directive in software is:
"Everyone did the best they could with the information they had at the time."
⬇️ It's called the "Prime Directive" - and it's not about Star Trek ⬇️
The prime directive in software is:
"Everyone did the best they could with the information they had at the time."
November 6, 2025 at 3:02 PM
There's an idea in software that's stuck with me over the years:
⬇️ It's called the "Prime Directive" - and it's not about Star Trek ⬇️
The prime directive in software is:
"Everyone did the best they could with the information they had at the time."
⬇️ It's called the "Prime Directive" - and it's not about Star Trek ⬇️
The prime directive in software is:
"Everyone did the best they could with the information they had at the time."
Reposted by Alex Lau
Every great developer I've ever worked with has had one trait in common
But, it doesn't always look the same...
That trait: Obsession.
Some can't rest until the architecture feels clean.
Some dive deep into weird bugs just to understand them.
But, it doesn't always look the same...
That trait: Obsession.
Some can't rest until the architecture feels clean.
Some dive deep into weird bugs just to understand them.
November 4, 2025 at 3:04 PM
Every great developer I've ever worked with has had one trait in common
But, it doesn't always look the same...
That trait: Obsession.
Some can't rest until the architecture feels clean.
Some dive deep into weird bugs just to understand them.
But, it doesn't always look the same...
That trait: Obsession.
Some can't rest until the architecture feels clean.
Some dive deep into weird bugs just to understand them.
Every great developer I've ever worked with has had one trait in common
But, it doesn't always look the same...
That trait: Obsession.
Some can't rest until the architecture feels clean.
Some dive deep into weird bugs just to understand them.
But, it doesn't always look the same...
That trait: Obsession.
Some can't rest until the architecture feels clean.
Some dive deep into weird bugs just to understand them.
November 4, 2025 at 3:04 PM
Every great developer I've ever worked with has had one trait in common
But, it doesn't always look the same...
That trait: Obsession.
Some can't rest until the architecture feels clean.
Some dive deep into weird bugs just to understand them.
But, it doesn't always look the same...
That trait: Obsession.
Some can't rest until the architecture feels clean.
Some dive deep into weird bugs just to understand them.
Explaining technical issues to PMs or stakeholders is a balancing act.
Too much detail, and eyes glaze over.
Too little, and the impact gets lost.
I've learned (often the hard way) that it's not just about what's broken - it's about why it matters.
❌ "this API is throwing a 500"
Too much detail, and eyes glaze over.
Too little, and the impact gets lost.
I've learned (often the hard way) that it's not just about what's broken - it's about why it matters.
❌ "this API is throwing a 500"
October 28, 2025 at 2:05 PM
Explaining technical issues to PMs or stakeholders is a balancing act.
Too much detail, and eyes glaze over.
Too little, and the impact gets lost.
I've learned (often the hard way) that it's not just about what's broken - it's about why it matters.
❌ "this API is throwing a 500"
Too much detail, and eyes glaze over.
Too little, and the impact gets lost.
I've learned (often the hard way) that it's not just about what's broken - it's about why it matters.
❌ "this API is throwing a 500"
When you start a side project, what's usually your goal?
Are you trying to build a real app?
Or just exploring something new to see where it goes?
For me, side projects have always been a chance to play.
No deadlines. No roadmap. No pressure to ship. Just a space to experiment.
Are you trying to build a real app?
Or just exploring something new to see where it goes?
For me, side projects have always been a chance to play.
No deadlines. No roadmap. No pressure to ship. Just a space to experiment.
October 23, 2025 at 2:02 PM
When you start a side project, what's usually your goal?
Are you trying to build a real app?
Or just exploring something new to see where it goes?
For me, side projects have always been a chance to play.
No deadlines. No roadmap. No pressure to ship. Just a space to experiment.
Are you trying to build a real app?
Or just exploring something new to see where it goes?
For me, side projects have always been a chance to play.
No deadlines. No roadmap. No pressure to ship. Just a space to experiment.
Reposted by Alex Lau
One thing I miss from not having worked with Rails for a while:
With Rails, you can spin up a new app in minutes with so much already wired up.
Auth, DB, structure, patterns - it's all just there.
With Rails, you can spin up a new app in minutes with so much already wired up.
Auth, DB, structure, patterns - it's all just there.
October 21, 2025 at 2:03 PM
One thing I miss from not having worked with Rails for a while:
With Rails, you can spin up a new app in minutes with so much already wired up.
Auth, DB, structure, patterns - it's all just there.
With Rails, you can spin up a new app in minutes with so much already wired up.
Auth, DB, structure, patterns - it's all just there.
One thing I miss from not having worked with Rails for a while:
With Rails, you can spin up a new app in minutes with so much already wired up.
Auth, DB, structure, patterns - it's all just there.
With Rails, you can spin up a new app in minutes with so much already wired up.
Auth, DB, structure, patterns - it's all just there.
October 21, 2025 at 2:03 PM
One thing I miss from not having worked with Rails for a while:
With Rails, you can spin up a new app in minutes with so much already wired up.
Auth, DB, structure, patterns - it's all just there.
With Rails, you can spin up a new app in minutes with so much already wired up.
Auth, DB, structure, patterns - it's all just there.
Reposted by Alex Lau
Some of the biggest lessons in my career didn't come from what I learned -
They came from what I had to unlearn.
Here are a few that stand out:
❌ You have to know everything.
They came from what I had to unlearn.
Here are a few that stand out:
❌ You have to know everything.
October 14, 2025 at 2:03 PM
Some of the biggest lessons in my career didn't come from what I learned -
They came from what I had to unlearn.
Here are a few that stand out:
❌ You have to know everything.
They came from what I had to unlearn.
Here are a few that stand out:
❌ You have to know everything.
Some of the biggest lessons in my career didn't come from what I learned -
They came from what I had to unlearn.
Here are a few that stand out:
❌ You have to know everything.
They came from what I had to unlearn.
Here are a few that stand out:
❌ You have to know everything.
October 14, 2025 at 2:03 PM
Some of the biggest lessons in my career didn't come from what I learned -
They came from what I had to unlearn.
Here are a few that stand out:
❌ You have to know everything.
They came from what I had to unlearn.
Here are a few that stand out:
❌ You have to know everything.
Reposted by Alex Lau
What's one of the most under-appreciated factors in software estimates?
Code familiarity.
Two tasks might look the same on paper - same complexity, same output. But one lives in code you've touched 20 times, whereas the other would be brand new code in your app.
Code familiarity.
Two tasks might look the same on paper - same complexity, same output. But one lives in code you've touched 20 times, whereas the other would be brand new code in your app.
October 9, 2025 at 2:04 PM
What's one of the most under-appreciated factors in software estimates?
Code familiarity.
Two tasks might look the same on paper - same complexity, same output. But one lives in code you've touched 20 times, whereas the other would be brand new code in your app.
Code familiarity.
Two tasks might look the same on paper - same complexity, same output. But one lives in code you've touched 20 times, whereas the other would be brand new code in your app.
Reposted by Alex Lau
There's a new kind of context switch in our workflow now:
Waiting for the LLM to respond.
You hit enter.
You wait.
You alt-tab.
Just a quick glance at Slack or email. Maybe review a PR. Maybe scroll for a second.
Then the model finishes. But your brain? It's already somewhere else.
Waiting for the LLM to respond.
You hit enter.
You wait.
You alt-tab.
Just a quick glance at Slack or email. Maybe review a PR. Maybe scroll for a second.
Then the model finishes. But your brain? It's already somewhere else.
October 7, 2025 at 2:05 PM
There's a new kind of context switch in our workflow now:
Waiting for the LLM to respond.
You hit enter.
You wait.
You alt-tab.
Just a quick glance at Slack or email. Maybe review a PR. Maybe scroll for a second.
Then the model finishes. But your brain? It's already somewhere else.
Waiting for the LLM to respond.
You hit enter.
You wait.
You alt-tab.
Just a quick glance at Slack or email. Maybe review a PR. Maybe scroll for a second.
Then the model finishes. But your brain? It's already somewhere else.
There's a new kind of context switch in our workflow now:
Waiting for the LLM to respond.
You hit enter.
You wait.
You alt-tab.
Just a quick glance at Slack or email. Maybe review a PR. Maybe scroll for a second.
Then the model finishes. But your brain? It's already somewhere else.
Waiting for the LLM to respond.
You hit enter.
You wait.
You alt-tab.
Just a quick glance at Slack or email. Maybe review a PR. Maybe scroll for a second.
Then the model finishes. But your brain? It's already somewhere else.
October 7, 2025 at 2:05 PM
There's a new kind of context switch in our workflow now:
Waiting for the LLM to respond.
You hit enter.
You wait.
You alt-tab.
Just a quick glance at Slack or email. Maybe review a PR. Maybe scroll for a second.
Then the model finishes. But your brain? It's already somewhere else.
Waiting for the LLM to respond.
You hit enter.
You wait.
You alt-tab.
Just a quick glance at Slack or email. Maybe review a PR. Maybe scroll for a second.
Then the model finishes. But your brain? It's already somewhere else.
Reposted by Alex Lau
The best developers I've worked with - the ones closest to the mythical "10x" - weren't just great at writing code.
They were great at not writing it.
They knew when to say:
- "We don't need this."
- "There's a library for that."
- "Let's solve the real problem, not overengineer the symptom."
They were great at not writing it.
They knew when to say:
- "We don't need this."
- "There's a library for that."
- "Let's solve the real problem, not overengineer the symptom."
October 2, 2025 at 2:03 PM
The best developers I've worked with - the ones closest to the mythical "10x" - weren't just great at writing code.
They were great at not writing it.
They knew when to say:
- "We don't need this."
- "There's a library for that."
- "Let's solve the real problem, not overengineer the symptom."
They were great at not writing it.
They knew when to say:
- "We don't need this."
- "There's a library for that."
- "Let's solve the real problem, not overengineer the symptom."
The best developers I've worked with - the ones closest to the mythical "10x" - weren't just great at writing code.
They were great at not writing it.
They knew when to say:
- "We don't need this."
- "There's a library for that."
- "Let's solve the real problem, not overengineer the symptom."
They were great at not writing it.
They knew when to say:
- "We don't need this."
- "There's a library for that."
- "Let's solve the real problem, not overengineer the symptom."
October 2, 2025 at 2:03 PM
The best developers I've worked with - the ones closest to the mythical "10x" - weren't just great at writing code.
They were great at not writing it.
They knew when to say:
- "We don't need this."
- "There's a library for that."
- "Let's solve the real problem, not overengineer the symptom."
They were great at not writing it.
They knew when to say:
- "We don't need this."
- "There's a library for that."
- "Let's solve the real problem, not overengineer the symptom."
Reposted by Alex Lau
In software, you don't get extra credit for building everything from scratch.
Unless it's your differentiator, don't reinvent the wheel.
I've seen teams roll their own auth, config systems, feature flags, even analytics platforms — all because they convinced themselves that they should.
Unless it's your differentiator, don't reinvent the wheel.
I've seen teams roll their own auth, config systems, feature flags, even analytics platforms — all because they convinced themselves that they should.
September 30, 2025 at 2:01 PM
In software, you don't get extra credit for building everything from scratch.
Unless it's your differentiator, don't reinvent the wheel.
I've seen teams roll their own auth, config systems, feature flags, even analytics platforms — all because they convinced themselves that they should.
Unless it's your differentiator, don't reinvent the wheel.
I've seen teams roll their own auth, config systems, feature flags, even analytics platforms — all because they convinced themselves that they should.
In software, you don't get extra credit for building everything from scratch.
Unless it's your differentiator, don't reinvent the wheel.
I've seen teams roll their own auth, config systems, feature flags, even analytics platforms — all because they convinced themselves that they should.
Unless it's your differentiator, don't reinvent the wheel.
I've seen teams roll their own auth, config systems, feature flags, even analytics platforms — all because they convinced themselves that they should.
September 30, 2025 at 2:01 PM
In software, you don't get extra credit for building everything from scratch.
Unless it's your differentiator, don't reinvent the wheel.
I've seen teams roll their own auth, config systems, feature flags, even analytics platforms — all because they convinced themselves that they should.
Unless it's your differentiator, don't reinvent the wheel.
I've seen teams roll their own auth, config systems, feature flags, even analytics platforms — all because they convinced themselves that they should.
Reposted by Alex Lau
Some of the most useful things I've learned didn't come from work.
They came from side projects.
Hackathons, experiments, late-night curiosities — they've all taught me something I didn't expect.
They came from side projects.
Hackathons, experiments, late-night curiosities — they've all taught me something I didn't expect.
September 25, 2025 at 2:07 PM
Some of the most useful things I've learned didn't come from work.
They came from side projects.
Hackathons, experiments, late-night curiosities — they've all taught me something I didn't expect.
They came from side projects.
Hackathons, experiments, late-night curiosities — they've all taught me something I didn't expect.