Companies will give you a bug and say, "Fix it in one hour."
No time to think. No time to test under load. No time to verify if your fix actually holds up.
Just rush and ship.
Back then? We'd fix a bug and spend another hour making sure it survived traffic. We'd check every query. We'd think it through.
Now? Deadlines. Penalties. Stress.
And we wonder why software is breaking everywhere. 😩
#TechDeals #TechTips
Honestly? I think Lope nailed the core of it . AI doesn't make good code good or bad code bad. It just makes both faster.
But here's where I'd push back a little.
On disclosure:
Yes, the code is open source and people can look. But let's be real most people won't dig through your commits to figure out what was AI-generated. So it's not about obligation. It's about transparency and trust.
If you're proud of your work, why not just say, "Yeah, I used AI for parts of it here's what I owned and here's what I shipped"? That builds more trust than being vague.
On the "visitor's burden":
I don't think it's a burden. I think it's just curiosity. And if someone asks, answer honestly. It takes 2 seconds. Dodging it makes it seem like you're hiding something — even if you're not.
On quality:
Lobe said it best "The code either works or it doesn't." That's the only metric that actually matters at the end of the day.
Bottom line for me:
AI is a tool. Use it. Own it. Be honest about it. And don't let anyone discredit your work just because you used it but also don't give them a reason to.
Fellow devs,
Let's be real for a second.
You can be the best engineer on the team. The one who holds things together. The one everyone runs to when things break.
But if the company decides they want you out, you're out. Trust me, they've got the money, the power, and the plans to survive without you.
It hurts. But it's true.
So what's the move?
Build options. Save money. Learn new skills. Create things on the side. Make friends outside your workplace.
Don't wait until you're replaced to realize you should have prepared.
Start today.
AI can code now, so why should I learn?
Here's the hard truth: If you don't understand how code works, you won't even know if the AI is giving you garbage. You'll be at its mercy.
Learn to code so you can control the AI not the other way around.
You can vibe-code your entire backend no doubt about that.
But DevOps? That's a different beast.
You can't vibe your way through infrastructure configurations without manual oversight. Automation is great, but without careful control, you'll burn through cloud costs, rack up unexpected bills, and end up paying for mistakes you didn't even see coming.
Vibe-coding works for features. DevOps needs discipline.
Let's be real most of us don't read those long "my experience" posts. We scroll. We like. We move on.
But what if we recorded ourselves instead? Told our stories the way they're meant to be told with emotion, tone, and real expression. That's what I'm doing here.
No walls of text. Just me sharing my experience like we're having a conversation.
Watch, listen, and tell me do you prefer videos or write-ups?
Why doesn't SQL have an "Undo" button? 😩
You forget one tiny column. You hit execute. And just like that, it's done. No pop-up asking "Are you sure?" No second chances. Just a quiet, ruthless confirmation that your data is now… different.
The worst part? You can't edit what you just did. You can't hit Ctrl + Z and pretend it never happened. All you can do is stare at the screen and ask yourself, "What did I just do?"
And then the scramble begins — writing a reverse query, checking backups, praying you didn't just break production.
How many times have you wished for an Undo button in SQL? 🙃
If you've ever accidentally updated a table without a WHERE clause, this one's for you.
Synchronous is simpler to build but painful to scale.
Asynchronous is harder to set up but gives you a better user experience and more resilience.
So as a backend engineer, you don't pick one over the other — you pick what fits the use case.
Had a genuinely confusing one recently marking a trade as paid would randomly fail with "trade not found." Same exact trade, same exact code, sometimes it just... didn't work.
Turned out our tenants can have multiple accounts connected on the exchange we work with, and somewhere in our code we had a query fetching "the account's token" without telling the database which one to pick when there were multiple matches. So the database just picked one at random, basically. If it happened to grab the wrong account, the exchange (correctly) said it couldn't find the trade because from that account's point of view, it really couldn't.
Nothing was wrong with our actual trade-handling logic. We'd just left a decision up to chance without realizing it.
Fixed it by always resolving the correct account directly from the trade itself, instead of guessing based on "the tenant."
Funny (and slightly humbling) lesson: if you don't tell your database how to sort results, it's not giving you a sensible default it's rolling dice.
#fypviralシ #fypシ #viralreelschallenge #viralvideochallenge #contentcreator
If your team doesn't understand your goal, they can't execute. Period.
It sounds simple, but you'd be surprised how many founders overlook this. You assume everyone sees the bigger picture the way you do but they don't. They're not in your head. They don't attend every strategy meeting you have with yourself at 2 AM. All they see is the task in front of them.
And when all they see is a task, they treat it like a checkbox. They complete it. They move on. But they don't connect it to the mission. They don't feel the weight of why it matters. And when that happens, the quality drops, the motivation fades, and suddenly you're wondering why your team isn't performing at the level you expected.
Here's the hard truth: execution is not about skill alone. It's about alignment.
You can have the smartest engineers, the sharpest marketers, the most experienced operators but if they don't know where the ship is headed, they'll row in different directions. And no matter how fast they row, you're not going anywhere.
So as a founder, your job isn't just to set the goal. It's to communicate it relentlessly in meetings, in slack messages, in one-on-ones, in casual conversations. Repeat it until you're tired of hearing yourself say it. Because the moment you stop talking about it, your team stops thinking about it.
And the moment they stop thinking about it, they stop executing.
Don't let that happen.
#tech
#TechDeals
#founders