Maxl0nG

Exploring the future of AI 🚀 | Sharing startups, tools & practical ideas | How to make money with AI | DM for collabs

[Maxl0nG @MaxL0nG]

https://t.co/Jk4kTvUJxd

[Maxl0nG @MaxL0nG]

@ajs6888 我部分同意你的看法。做一个 demo 其实不难，用 YOLO，再加上一些传感器和摄像头，就能验证整个方案是否可行。真正难的是落地：要把电子系统集成起来，保证整个系统稳定、可靠地运行，更重要的是，还要有合适的实际应用场景。这才是真正的挑战

[Maxl0nG @MaxL0nG]

You can build LaserWeeds with Claude API in 2 weeks.
Claude's capabilities:
 Vision API — analyze plant images
 Reasoning — classify weeds vs crops
 Integration — connect to hardware (laser)
💰Cost: $0.003 per image analysis
What it takes:
•1-2 engineers
•Training dataset: 1,000 labeled images
•Deploy on NVIDIA Jetson
•Total dev time: 14 days
Result: $2,400 savings/season for farmers.
Claude turned weed detection from impossible to doable. This is how AI startups actually get built now.

[Maxl0nG @MaxL0nG]

I gave Claude Code a memory that survives context resets, session ends, and compaction.

5 hooks. No manual re-explaining. Ever again. https://t.co/T4XG64B6Od

[Maxl0nG @MaxL0nG]

@Bober_smart and people also look at her for her beautiful eyes😂

[Maxl0nG @MaxL0nG]

@AgentBishop111 Well, at least there are already lasers that shoot down mosquitoes😁

[Maxl0nG @MaxL0nG]

@cryptowluha By the way, this old man is not as smart as everyone thinks, he has a lot of contradictions

[Maxl0nG @MaxL0nG]

@0xMavex In 5 years, the Terminator movie won't be science fiction.

[Maxl0nG @MaxL0nG]

@VanllonG I think it has to do with certain sensors, but I don't understand why it doesn't affect other plants.

[Maxl0nG @MaxL0nG]

A Fisherman Invented a Way to Catch Any Fish Using Artificial Intelligence

A passionate fisherman has turned his hobby into an innovative technology project by combining artificial intelligence with modern computer vision. Using Claude, he developed an application that can recognize fish in real time and instantly provide detailed information about them.
0:12

The app analyzes video from a camera, identifies the species of fish, and displays essential details such as where it lives, what it eats, the best time of day to catch it, and the most productive fishing season. The system uses advanced computer vision techniques, and when needed, integrates the YOLO object detection model for highly accurate fish recognition, even in challenging environments.

Beyond identifying fish species, the application generates a complete information card with recommended bait, feeding habits, peak fishing times, seasonal activity, and additional facts about the species. This allows anglers to make smarter decisions directly at the water and significantly improve their chances of a successful catch.

According to the project's creator, the goal is to make fishing easier and more effective for everyone. With the help of artificial intelligence, even beginners can quickly identify a fish, learn the best strategy to catch it, and understand when and where they are most likely to succeed.

This project demonstrates how AI and modern technology can transform even a traditional hobby like fishing, turning a smartphone into a powerful digital fishing assistant.

[Maxl0nG @MaxL0nG]

https://t.co/Jk4kTvUJxd

[Maxl0nG @MaxL0nG]

He sat through a YOLO conference talk. Now his pig farm will never be the same

He wasn't even supposed to be that into it. Just another conference, another talk about computer vision — until someone on stage showed a live demo: a camera reading everything in frame in real time, no manual counting, no guesswork.

He sat there thinking about his farm. Dozens of pigs, and the only way to know a single one's weight is to physically drag it onto a scale — a whole production every time, for every animal, over and over.

And then it clicked: if a camera can read that much just by looking, why is he still hauling pigs onto a scale by hand?

So he opened up Claude and started building. No computer vision background, no team, just an idea and a conversation. A few days later he had it: point a camera at a pig, and it scans it — instantly.

He didn't build a startup. He built the tool he actually needed, in an afternoon, by describing what he wanted out loud.

This is what's coming. Not "someday" — now. If you can describe what you need, you can build it.

[Maxl0nG @MaxL0nG]

https://t.co/Jk4kTvUJxd

[Maxl0nG @MaxL0nG]

@vint_98 Very good video with I will save it😅

[Maxl0nG @MaxL0nG]

From tracking cement thieves to tracking chickens — one YOLO

A guy worked as a security guard at a cement factory. Night shifts, a wall of cameras, and part of his job was making sure nobody walked off with bags of cement. The factory ran a system that tracked every movement of the warehouse staff through the cameras — who came in, who picked something up, who carried something out. Not a single bag slipped by unnoticed.

Meanwhile, back home, he kept chickens. And that was total chaos — a chicken would go missing, or he'd have no idea how many were even left after the neighbor's dog got loose near the fence. Counting them by hand was a gamble, especially with fifteen of them running around at once.

Then one night on shift, staring at the screen as the system tracked every move in the warehouse, it hit him: why not build the exact same thing — for the chickens?

He grabbed YOLO, trained it on his chickens, added tracking so each one got its own ID and never got counted twice — and ended up with a bot that shows, in real time, how many chickens are in frame right now and how many passed through that day.

Now his chickens barely go missing without him noticing — he sees every single movement, just like that system at the warehouse, except instead of cement bags, it's his chickens 🐔

The factory counted bags. Now he counts chickens.

[Maxl0nG @MaxL0nG]

@kojrey_codes As an extra tool, I don’t see why not. At the very least, it could help with counting sheep.
Thanks for reading!🤝

[Maxl0nG @MaxL0nG]

He stopped losing sheep with a $60 camera — and made $5,000+ a year off it.

He was losing 5-10 sheep a month. Wolves hit at night, he found out in the morning — money already gone.
At ~$150-200 per sheep, that's over $3,000 a year walking off into the woods.

He'd never written a line of code. So he asked Claude to build the whole thing — the detection program that spots and counts every sheep, plus the alert system behind it.

The setup: a camera that counts sheep every 5 seconds and pings Telegram the moment the count drops. Claude wrote the code end to end — he just described the problem.
Not a $5,000 thermal camera. Not a $200/month security subscription. A regular $60 webcam and a model that runs locally — no monthly bills.
Built it in a weekend, entirely with Claude writing the code.

Result after 2 months: 0 sheep lost. $3,000+ a year now stays in his pocket instead of feeding wolves — for a one-time $60.
Word got around. Other farmers saw it working and started asking for the same setup.

He's already sold the app to other farmers — $5,000 in sales, from a tool Claude helped him build to solve his own problem

[Maxl0nG @MaxL0nG]

https://t.co/Jk4kTvUJxd

[Maxl0nG @MaxL0nG]

@JackYuan35 But a single wolf can wipe out an entire flock of sheep, causing huge losses

[Maxl0nG @MaxL0nG]

@Asia_Lazy_Sheep I like the way you think, but I don’t understand why you’d need to monitor vegetables and other produce like that.

[Maxl0nG @MaxL0nG]

https://t.co/Jk4kTvUJxd

[Maxl0nG @MaxL0nG]

@nickjmaurer I agree, but overall it’s interesting how someone managed to apply AI to a market that doesn’t seem like an obvious fit at first glance

[Maxl0nG @MaxL0nG]

@JackYuan35 But the wolf won’t get far, and it can be caught.

[Maxl0nG @MaxL0nG]

@jsonparse13 If the AI can detect a sheep, it can also detect a wolf and send a notification to the farmer

[Maxl0nG @MaxL0nG]

@nickjmaurer It's actually not that hard if you know what to do and provide a good prompt

[Maxl0nG @MaxL0nG]

@thec0d3z I work with AI too, and honestly, why couldn’t he do something like that? As far as I’m concerned, Claude is totally capable of doing it. Just look at the YOLO model.

[Maxl0nG @MaxL0nG]

@jsonparse13 Because sheep are needed on the farm for their wool