0:04
Someone converted their Steam library into SSD "cartridges", and you slot them in like a Game Boy
I stumbled upon this thing where someone took their Steam library and converted it into these custom SSD cartridges, kind of like the old Game Boy cartridges. They're actually just modified USB sticks, but they're designed to look and feel like those retro cartridges. Each one has a unique game installed on it, and you have to physically insert them into a custom dock for them to work. It's not just a novelty, either - the creator says it's actually faster and more convenient than loading games from a traditional SSD.
0:20
Even HP resellers thought the price of toner and ink was too high – so HP India facilitated an illegal cartel
HP’s Indian arm got slapped with a ₹138.85‑crore fine after the competition regulator found it had nudged its resellers into a coordinated pricing scheme. The CCI says HP told certain partners exactly what to quote on government tenders, and even barred some from bidding, to keep a tight grip on printer‑ink sales. WhatsApp chats revealed a pattern of cover bidding—one reseller inflating a price so the others could win with more realistic offers—plus pre‑assigned customer allocations from 2017 to 2020.
The regulator says HP’s motive wasn’t just protecting its market share; it was also trying to stop tier‑2 dealers from slipping into counterfeit ink to stay competitive. By fostering an “understanding” among those resellers, HP hoped to keep prices from spiralling down and preserve the legitimacy of its supply chain.
A second order looked at PC sales, where HP allegedly steered the reverse‑auction process to make sure at least one HP reseller stayed in the final round, avoiding a scenario where all partners bailed because of unsustainable price pressure.
All parties are now ordered to stop the collusive practices, and HP will have to absorb the monetary penalty without any further relief.
0:53
Chromatography as Art
You know how I've been digging into the intersection of art and science? Well, I stumbled upon this fascinating project that's got me thinking. It's all about using chromatography as a form of art, and I'm not just talking about the science behind it. The creator, Suchir, is using this ancient technique to separate mixtures and create these stunning, intricate patterns.
It's actually pretty simple – all you need is a sketch pen, some paper, a coffee filter, and a glue stick. The magic happens when you mix different substances, like black ink and water, and watch as they separate into their constituent pigments. It's like a butterfly effect, where the outcome is completely unexpected and beautiful.
What I love about this project is that it's accessible to anyone, regardless of their background or interests. It's a great way to explore the intersection of art and science, and it can even be used to teach kids about chemistry and the properties of different materials. Plus, it's a great way to spark creativity and imagination.
One of the coolest things about this project is that it can be used for more than just art. For example, you can use it to measure the caffeine content in different types of coffee or tea. It's a great way to apply scientific principles to real-world problems, and it's a lot of fun to boot.
1:28
You can now build AI agents that order DoorDash directly from the command line
So, I was digging into this DoorDash CLI beta and it's actually pretty interesting. They're allowing developers to build AI agents that can order food directly from the command line. I mean, think about it, you could technically automate ordering food for yourself or even for others.
It's still in beta, but you can sign up for the waitlist if you're a macOS developer in the US or Canada. The fact that they're targeting developers specifically is a good sign, it suggests they're serious about making this a robust tool.
I'm curious to see how this will play out, especially with the rise of AI agents like Claude. It's not hard to imagine a future where you can just tell your AI assistant to pick up dinner on the way home.
1:48
Cyberattack threatens utterly critical infrastructure in Japan: KFC
Nichirei Group, the Japanese frozen‑food and cold‑logistics firm, confirmed a cyberattack that knocked out the servers handling its warehouse shipments and personal data. The breach crippled the flow of ingredients to KFC Japan, forcing the fast‑food chain to pause orders on its app and website and warn that some stores might trim menus or even close temporarily.
Because the attackers accessed a server holding customer information, Nichirei is keeping details tight, hoping to avoid giving the culprits any clues. The company says it aims to get systems back online by Friday, but the exact scope of the disruption remains unclear.
KFC hasn’t announced specific closures yet, but the incident highlights how a single breach can ripple through supply chains, affecting everything from logistics to the meals we expect on a hot summer day.
2:12
US Suffered a Major Power Outage Every Month of 2026
I dug into the claim that the U.S. saw a major blackout every month in 2026 and the pattern that emerged was oddly consistent: each month a different weather event knocked out hundreds of thousands of customers. In January a freeze left nearly 300 k without power, February’s snowstorm hit the Northeast with 380 k dark, March’s Midwest storms pushed the count past a million, and April’s tornado outbreak in Illinois added more. May saw the Energy Secretary issue an emergency order for backup generation during a heatwave, while June’s extreme weather again left over 370 k in the dark. The list keeps growing, and while “major” is subjective, the sheer frequency of weather‑driven outages makes the statistic feel almost inevitable. It’s a reminder that each event is a personal story for the people left without lights, even if the headlines gloss over the human side.
2:35
Three insights you may have missed from theCUBE’s coverage of RAISE Summit
Hey, I just got done reading through theCUBE's coverage of RAISE Summit and I wanted to share a few things that caught my attention. So, it looks like agentic inference is changing the way we think about AI infrastructure. Instead of just scaling up training, we're now talking about expanding context windows and memory-augmented reasoning. That's a big shift from where we were before.
The thing that really stood out to me was how graphics processing units (GPUs) are being used. Apparently, they need to be continuously fed with data in order to keep up with the demands of agentic systems. That's not something you hear about every day. It's like the GPUs are being asked to do a lot more than just crunch numbers.
Storage is also becoming a major player in this new landscape. As enterprises dive deeper into agentic systems, they're realizing that storage needs to be able to keep up. It's not just about storing data, it's about being able to access and process it quickly. That's a pretty significant challenge.
I also found it interesting that the focus is shifting from just scaling up training to actually understanding how these agentic systems work. It's like we're moving from a "how big can we make it?" mentality to a "how does it actually work?" mentality. That's a more nuanced and mature approach to AI development.
Overall, it feels like we're entering a new phase in the development of AI infrastructure. It's not just about throwing more resources at the problem, it's about understanding the underlying mechanics and being able to adapt to changing demands.
3:17
Pinch Puts an Arduino On a USB-C Connector
The pinch is a tiny 32‑bit board that literally sits on a USB‑C plug. Under the hood is a SAMD11 Cortex‑M0+ running at 48 MHz with 4 KB SRAM and 16 KB flash, so you lose a bit of headroom compared to an ESP32 but still get native USB and 15 GPIOs, plus the usual SPI, I²C, PWM and UART. Because the bootloader eats a quarter of the flash, you’re left with about 12 KB for your sketch, which feels tight but works for simple sensor or button projects.
What’s surprising is how they’ve packaged all that into a footprint the size of a female USB‑C connector, and they even provide a tiny breakout board to make wiring manageable. The design files and a 3‑D STEP model are already out, so you can start drafting your own tiny enclosures.
They’re in the final testing phase, with pre‑orders open at $16 and shipping slated for September. If you need something that fits on the tip of a finger and can still talk USB, this might be the smallest complete solution you’ll find right now.
3:43
Former OpenAI CTO does what Altman won't: releases a frontier AI model that's actually open
If you’re in the market for a frontier-class open weights model, your options are few and far between outside of the Chinese model houses. With the Wednesday release of a new model code-named "Inkling", an outfit called Thinking Machines Lab aims to change that. Founded in early 2025 by former OpenAI CTO Mira Murati, Thinking Machines' first model is a big one. Weighing in at 975 billion parameters, the model requires more than two terabytes of GPU memory — a quantity present in around eight of Nvidia's B300 accelerators, or sixteen H200s — to run at its native 16-bit precision.
4:00
Did AI decide who lost their jobs? Meta is heading to court over that question
Enterprises that use AI in hiring and firing decisions continue to be under scrutiny, and this time it’s Meta under the microscope.
A legal complaint filed on July 13 in a US District Court in California alleges that Meta used AI systems that unfairly and illegally selected workers for termination while they were out on protected leave.
More than two dozen anonymous plaintiffs are seeking a preliminary injunction that would prevent the company from finalizing their separations or altering their compensation, benefits, or protected leave status.
Meta has countered that the claims lack merit
