Pinned Tweet
I gave an AI coding agent a real project and one simple outcome:
Understand the project, research similar memory-agent projects, and turn the findings into a Word document.
No step-by-step instructions.
I wanted to see if Qoder could actually handle the work.
🤝 Paid partnership
HarriStack retweeted
For 70 years, we were told plastic was safe because it doesn't react with anything.
Nature just published proof that it does. And the trigger is water.
Researchers at Columbia University and UC Berkeley took six of the most common plastics on Earth.
Polyethylene. Polypropylene. Polystyrene. PVC. PET. Acrylic.
The stuff in your water bottles, food containers, pipes and packaging.
They ground them into microplastics and dropped them into plain water.
Nothing else. No sunlight. No chemicals. Just plastic and water.
And the water started changing.
At the surface of every single plastic, it began producing hydroxyl radicals. One of the most reactive and destructive molecules in chemistry.
All six plastics did it. Every one.
To prove it, they added a yellow dye to water with polystyrene. Over 10 days, nearly 90% of the dye was destroyed. The water turned almost clear.
In other liquids, the dye survived. Only water made the plastic active.
In plain English:
The plastic isn't leaking a toxic chemical.
It is turning the water around it into something that attacks.
Then they tested what that does to living things.
DNA and proteins got pulled onto the plastic surface like magnets.
Proteins started clumping together.
And red blood cells that touched the particles were torn into broken fragments.
But here is the twist.
The most dangerous plastics were the ones we trusted most.
Polyethylene and polypropylene, the "safe" plastics in bottles and food containers, were the most reactive of all.
The effect was weaker in blood serum. But it was still there.
And microplastics have already been found in human blood, brains, placentas and arteries.
The lead researcher said it plainly: "Water is a key active player here, rather than just a spectator."
We spent decades asking what chemicals leak out of plastic.
Nobody thought to ask what plastic does to the water inside us.
How many times have you sipped from a plastic bottle today?
HarriStack retweeted
Create the visuals. Cut the clips. Get your brand out there. 🔥
Dreamina + CapCut PC makes a handy marketing workflow, and the combo starts at $6.29. Offer ends Oct 30, so don't leave it too late.
#CapCutDidThat #Dreamina #Smallbusinessmarketing #CapCutSale
HarriStack retweeted
Your content budget just got a little breathing room. 💸
Save up to 80% with the Dreamina + CapCut Pro offer, starting at just $6.29 combined.
Generate your marketing assets in Dreamina and edit them in CapCut PC.
🔥 Limited-time savings: Oct 20–30 on CapCut Desktop only.
#CapCutDidThat #Dreamina #Smallbusinessmarketing #CapCutSale
HarriStack retweeted
🚨Someone just made 70B parameter models run on a single 4GB GPU.
An open source tool called AirLLM has changed how large language models run on consumer hardware. Instead of loading the entire model at once, it processes one layer at a time. This means your VRAM needs depend only on the size of a single layer rather than the full model.
No quantization.
No distillation.
No pruning.
100% Open Source. Apache 2.0 License.
→ DeepSeek V3 (671B) runs on just 12GB
→ Llama 3.1 405B runs on 8GB
→ Kimi K3 (2.8 trillion parameters) runs on under 4GB
→ Works with nearly every major open source model
The largest model often needs the least VRAM. Kimi K3 is a sparse mixture of experts model, so it only loads the specific experts activated by each token instead of an entire dense layer. This is why a 2.8 trillion parameter model can run in less memory than a 70B model.
AirLLM makes frontier scale AI practical on everyday hardware without any of the usual tradeoffs.
(Repo in the comments)
HarriStack retweeted
CapCut’s Black Friday savings are now available to claim: 60% off Pro and 40% off Ultra, October 20–30, exclusively on CapCut Desktop.
The main product shot deserves a careful review in any Halloween campaign.
Local Inpainting in CapCut PC gives you a way to revisit a selected area of an AI-generated shot. Refine a detail and assess the updated image within the sequence, where you can judge its contribution to the overall ad.
Keep developing the visuals that carry the campaign’s central idea.
#capcutpc #capcutmarketing #capcutsale #capcutblackfriday
HarriStack retweeted
Today we’re announcing two new compute types on Shadeform:
CPUs, which are available today, and Apple Silicon, which is planned to release later this quarter 👇
The Uncomfortable Truth About the Tesla Roadster’s 9-Year Delay
Tesla promised the Roadster in 2017, with deliveries initially expected in 2020. The timeline kept shifting, leading to years of skepticism and vaporware accusations.
When the car was revealed on October 15, 2026, it appeared far more ambitious than the original prototype.
The 2017 prototype promised 1.9-second 0–60 acceleration. The new version claimed 0.81 seconds. The latest reveal also showcased SpaceX thrusters, a suction system, robot-inspired actuators, and aerospace-grade materials.
**His final point:**
You watched the Roadster fly, saw the backflip, and shared the 0.81-second acceleration clip. Then the internet moved on.
But beneath the spectacle are 11 engineering features that deserve attention, including the suction system, thrusters, retractable headlights, high-speed charging, and active aerodynamics.
The Performance version is expected to offer many of these technologies without the flying capabilities.
The spectacle was the flip. The real story is the engineering beneath it.
If this changes how you see the Roadster reveal, repost the first post so others can discover the engineering details behind the spectacle.
From the suction fan system and thrusters to the pop-down headlights, robot-actuated wing, 500 kW charging, and steer-by-wire controls, there’s more to the Roadster than its viral backflip.
Follow @ HarrisDecodes for breakdowns of hidden engineering, overlooked details, and breakthroughs in technology that most people miss.










