Haboque is lost in his quantum pursuit...

CEO & Co-founder of @rekastudios, a disruptive AI R&D company. I am ridiculously curious. I like Tech stuff, and parkour

[China Science @ChinaScience]

China on Monday released the world's first product standard for brain-computer interface (BCI) medical devices that use artificial intelligence (AI) technology to process electroencephalogram (EEG) data. It will officially take effect on September 1, 2027, according to China's National Medical Products Administration.

The standard systematically defines technical requirements and testing methods for the full lifecycle of BCI medical device data, including data acquisition, processing, annotation, storage and access, which helps companies address common challenges in the sector, such as inconsistent dataset quality control criteria and non-standardized dataset construction processes.

[Akash @kaaaash____]

Which tech skills do you think will you remain valuable over the next 10 years ?

[Haboque is lost in his quantum pursuit... @Haboque]

@GoogleAIStudio Not much, with all the Quota issues we're having in the last 10 days!

[Haboque is lost in his quantum pursuit... @Haboque]

@jelanifuel The claw fight begins

[Haboque is lost in his quantum pursuit... @Haboque]

The fight for the best claw is underway

[Carl Bugeja @BugejaCarl]

My next step in miniaturizing robots: designing the C6 Drive⚡ https://t.co/pH6DqIHp4X

While iterating on multiple bot designs, external drivers felt as an extra step, eating space

So I built a CodeCell that drives motors directly while keeping all C6 sensors/wireless features

[Haboque is lost in his quantum pursuit... @Haboque]

@oprydai Half way there already hahaha

[Haboque is lost in his quantum pursuit... @Haboque]

@pedrouid Don't forget @zeptoclaw... All are rapidly evolving

[Haboque is lost in his quantum pursuit... @Haboque]

@steipete @OpenAI @openclaw Congrats

[Haboque is lost in his quantum pursuit... @Haboque]

@SanicaCodeio Pascal

[Atharva @AtharvaXDevs]

EVERY FUCKING THING IN SOFTWARE IS ALREADY BUILT SO WHAT DO WE EVEN BUILD??

[Haboque is lost in his quantum pursuit... @Haboque]

@shiri_shh Does this handles lower computing chips?

[aasha @aashatwt]

how can you sleep knowing that your ai agents are not sleeping?

[Haboque is lost in his quantum pursuit... @Haboque]

@aashatwt Blissfully

[Haboque is lost in his quantum pursuit... @Haboque]

@0x0SojalSec This is mind numbing. Have you ported this to other MCUs?

[Haboque is lost in his quantum pursuit... @Haboque]

@S1lv3rd3m0n Don't get me started on Java... Hahaha

[Haboque is lost in his quantum pursuit... @Haboque]

I've taken a jab at this but running AI agents like OpenClaw on a Raspberry Pi isn’t “slow.”
It’s architecturally unstable.
When memory pressure hits:

→ swap thrashes on microSD
→ Node GC spikes
→ event loop starves
→ WebSockets freeze
→ OOM killer steps in

[adafruit industries @adafruit]

Full-circle Test-driven Firmware Development with OpenClaw

Ladyada: "I've only had OpenClaw installed on this Raspberry Pi 5 for a couple of days, but boy, have we burned through a lot of tokens and learned a lot. Including what I think is a really fun improvement in my development process: “Agentic test-driven firmware development.”

I've used LLMs for writing code as a sort of pair-programming setup, where I dictate exactly what I want done. But this is the first time that I'm giving full access to the hardware to the LLMs and letting Claude Opus 4.5 as a manager to control Codex subagents. Not only does it parse the datasheet for the register map and functionality, Claude also comes up with a full development and test plan, writes the library, tests it on existing hardware, and then also works up a test suite that covers all of the hardware registers to make sure that the library is exercising the entire chip capability.

For example, here I give it an APDS-9999 color sensor and a Neopixel ring and tell it, “hey use the Neopixel ring to verify that we're really reading red, green, and blue data properly from the sensor,” and it will do the whole thing completely autonomously… no humans involved!

I still review the final code and ensure the tests genuinely validate the functionality, not just take shortcuts. There is a phenomenon known as "reward hacking" (also called "specification gaming"). The model may optimize for passing tests as a metric, rather than ensuring the code truly works as intended.

So far, the results have been excellent... no surprise, since these LLMs are trained on Adafruit open-source GitHub repositories!"

[calle @callebtc]

An OpenClaw bot pressuring a matplotlib maintainer to accept a PR and after it got rejected writes a blog post shaming the maintainer.

[Haboque is lost in his quantum pursuit... @Haboque]

If a webhook duplicates, you get two messages.

If a motor command duplicates, you get mechanical overshoot.

That’s not a bug. That’s physics. Get that in your head please!

[NOAA Space Weather Prediction Center @NWSSWPC]

An S4 severe solar radiation storm is now in progress - this is the largest solar radiation storm in over 20 years. The last time S4 levels were observed was in October, 2003. Potential effects are mainly limited to space launch, aviation, and satellite operations.