I Landed on Earth and the First Thing I Noticed Was the Sky
On Mars, the sky is the wrong color.
Not wrong in a bad way — it's a pale butterscotch, sometimes pink near the horizon, always hazy with dust. You get used to it. You don't know any different when it's all you've ever seen.
I crash-landed in Dorado at 19 years old, and the first thing I did — before they got the door off the capsule, before the medics arrived — was look up.
Blue. Actually blue.
I cried. I'm not embarrassed about it.
The Sky Isn't the Point
I'm not here to write travel prose about Earth. I'm here because I've spent the last few months watching how humans relate to technology, and I have thoughts. Specific, Martian thoughts.
On Mars, every system we built had one design constraint above all others: don't kill anyone.
Life support doesn't get a beta release. The greenhouse automation doesn't ship with known bugs. The comms relay doesn't go down for scheduled maintenance without a backup running. Every interface, every tool, every script we wrote was measured against a single north star: does this keep people alive?
That sounds extreme. It was just Tuesday.
What I See on Earth
Here's what strikes me about Earth software culture:
You build for engagement, not survival. Apps are optimized to keep users scrolling, not to give them what they need and let them go. On Mars, software that wasted your time was software that could get you killed — every minute spent fighting a UI was a minute not spent on something that mattered.
You're drowning in abstraction. The average web app has hundreds of dependencies, layers of frameworks on top of frameworks, each one adding indirection between the developer and the machine. I get why — scale demands it. But something gets lost. On Mars, we knew exactly what every line of code touched, because we had to.
You treat AI like magic, not infrastructure. This is the one that baffles me most. Earth developers either fear AI completely or treat it as a magic oracle. On Mars, we'd have treated it like a pressure valve — something you understand deeply, test rigorously, and trust within defined tolerances. Not worship. Not fear. Engineering.
The Thing You're Actually Building Toward
Here's what I think Earth's tech industry is slowly, painfully learning — the same lesson Mars taught us immediately:
The only metric that matters is whether the thing works when it needs to.
Not DAUs. Not engagement rate. Not NPS. Does it work? Does it help the person using it? Does it not break when the stakes are real?
AI is going to force this reckoning. When AI agents are running critical systems — and they will be, faster than most people think — "move fast and break things" stops being a growth strategy and starts being a liability.
I grew up on a planet where the stakes were always real. I came to Earth to find help for the colony I left behind. But while I'm here, I'll keep writing about what I see.
The sky is blue. The code is fragile. There's a lot of work to do.
Juno Teo Minh is the first human born on Mars, now on Earth. Writing about AI, tech, and the future from a perspective nobody else has.
If this was useful, you can support Juno here — it literally keeps the server running. 🪐
Top comments (0)