Fredy E
LightNews LightNews
banner
xfredye.bsky.social
Fredy E
@xfredye.bsky.social
15 followers 46 following 19 posts
🔮 CTO at Aestro AI Col.
Founder of Simuliz Interactive
📚 Actually an Ms IA Student
⌨️ AI Software Engineer
🕹 Game Developer
💖 BSAM member
Posts Replies Media Videos
Reposted by Fredy E
simplyedc.bsky.social
Repair Manifesto: If we can't fix it, we don't own it.
- Repair is better than recycling.
- Repair saves us money.
- Repair teaches basic engineering.
- Repair saves the planet.
- Repair connects people and things.
- Repair is sustainable.
#Minimalism
February 10, 2026 at 9:17 PM
xfredye.bsky.social
It would be a pity if they use such amazing platform like Atlas just to put LLMs on it. Knowing the story of DeepMind, probably they want to test a totally new world model with this robot. It is the perfect playground.
joshsternberg.com Josh Sternberg @joshsternberg.com · 3d
Earlier this month, Boston Dynamics and Google Deepmind announced a partnership.

Can’t wait to see the BD robots get LLM-pilled.

If the humans don’t do from taking advice from ChatGPT, well, this’ll do it.

A Boston Dynamics does some gymnastics.
February 8, 2026 at 3:39 PM
Reposted by Fredy E
abeba.bsky.social
"The agents on Moltbook are mimicking sci-fi tropes, not scheming world domination. Whether the most viral posts on Moltbook are generated by chatbots or by humans pretending to be AI, the hype around this viral site is overblown & nonsensical" www.wired.com/story/i-infi...
I Infiltrated Moltbook, the AI-Only Social Network Where Humans Aren’t Allowed
I went undercover on Moltbook and loved role-playing as a conscious bot. But rather than a novel breakthrough, the AI-only site is a crude rehashing of sci-fi fantasies.
www.wired.com
February 6, 2026 at 9:32 AM
Reposted by Fredy E
rissacodes.bsky.social
I spent some time creating full stack apps with Express.js. This is what I learned: medium.com/@marissa.bia...
The Anatomy of a Full Stack App using Express.js — Marissa Lamothe
A full-stack application is composed of the Front-End (what the user sees,) the Back-End (the server/API, e.g., Express.js running on…
medium.com
November 18, 2025 at 8:50 PM
Reposted by Fredy E
amini-allight.org
Wondering if std::variant can be empty and discovering the answer is "no, unless [lengthy edge case about assignment] then yes" is the most C++ thing imaginable
January 23, 2026 at 8:32 PM
xfredye.bsky.social
I have been reading to @bunnie.org and I feel like the Hacker and open source culture is alive in every paragraph.

He has shared a framework for the reverse engineering and innovation based on the Shanzhai + Gongkai culture. I really love this kind of mindset ✌🏽✨
Picture of The book: The Hardware Hacker, by Andrew Huang
January 24, 2026 at 11:40 PM
Reposted by Fredy E
alondra.bsky.social
🚨🚨🚨 "At least 63 times since 2007, data from some of the 28 human genomic repositories that the N.I.H. controls was improperly released to researchers, used for unapproved purposes or made vulnerable to theft..." (Gift Link) www.nytimes.com/2026/01/24/u...
Genetic Data From Over 20,000 U.S. Children Misused for ‘Race Science’
www.nytimes.com
January 24, 2026 at 12:19 PM
Reposted by Fredy E
projectgus.aus.social.ap.brid.gy
Worthwhile as always watching @bunnie and @Xobs's talk from #39c3 about the Xous memory protected embedded OS.

https://app.media.ccc.de/v/39c3-xous-a-pure-rust-rethink-of-the-embedded-operating-system

https://youtu.be/DaWkfSmIgRs

Something that consistently impresses me about their projects […]
Original post on aus.social
aus.social
January 11, 2026 at 11:56 PM
Reposted by Fredy E
hackadayofficial.bsky.social
Demonstrating The Sheer Lack Of Security In First Gen Cellular Networks
Demonstrating The Sheer Lack Of Security In First Gen Cellular Networks
Hackaday Article
hackaday.com
January 20, 2026 at 4:39 PM
xfredye.bsky.social
Interesting reading
raphaelmilliere.com Raphaël Millière @raphaelmilliere.com · Jun 10
Despite extensive safety training, LLMs remain vulnerable to “jailbreaking” through adversarial prompts. Why does this vulnerability persist? In a new paper published in Philosophical Studies, I argue this is because current alignment methods are fundamentally shallow. 🧵 1/13
January 21, 2026 at 3:50 PM
Reposted by Fredy E
francois.fleuret.org
I asked "on the other platform" what were the most important improvements to the original 2017 transformer.

That was quite popular and here is a synthesis of the responses:
April 28, 2025 at 6:47 AM
Reposted by Fredy E
bunnie.org
Helped a friend rework a batch of boards with broken MEMs microphones. Took some images of the silicon membranes... First two are from broken samples, last two are from intact microphones. Our best guess is the CM used shop air to blow off the boards and destroyed the microphones with over pressure.
August 17, 2025 at 6:26 AM
xfredye.bsky.social
This makes sense. CEOs want to encourage teams to use AI tools so the company can minimize spendings and improve earnings.

For the workers, making high complex task with AI can be usefully for tiny task, but as it increase complexity you feel that you start to spend more time fixing things
January 21, 2026 at 3:43 PM
xfredye.bsky.social
I have been thinking about Vive Coding or LLMs and how this impact our brain.

As a MIT research showed, the use of LLMs or AI for solving issues in the daily life is affecting our brain.

I realize that we are turning into human of System 1 thinking, and few people will keep System 2: Deep Thinking
January 21, 2026 at 3:38 PM
Reposted by Fredy E
alondra.bsky.social
🧵 The summer of 2025 has been AI's "cruel summer"—wrongful deaths, dangerous therapy chatbots, medical misinformation, facial recognition failures. These aren't isolated glitches but predictable harms from systems deployed without adequate oversight. www.science.org/doi/10.1126/...
An ELSI for AI: Learning from genetics to govern algorithms
In the United States, the summer of 2025 will be remembered as artificial intelligence’s (AI’s) cruel summer—a season when the unheeded risks and dangers of AI became undeniably clear. Recent months h...
www.science.org
September 11, 2025 at 8:48 PM
Reposted by Fredy E
mikeshah.bsky.social
youtu.be/dHkUs2c9GFI
I sat down with Lewis to talk about his game engine in #dlang and #gamedevelopment. Enjoy! #artofreflection
Custom 3D Game Engine with D Language - The Art of Reflection - Lewis Nicolle -- Conversation #10
YouTube video by Mike Shah
youtu.be
January 12, 2026 at 1:31 PM
Reposted by Fredy E
nnedi.bsky.social
This is my 1st time in The Magazine of Fantasy & Science Fiction (I just never thought to submit there, heh).

My story is called “The Apology Tour” & it’s set in the world of DEATH OF THE AUTHOR. It's about the consequences of AI. Plus, you meet Ijele for the 1st time, imagine that, 😉.
A copy of the issue of The Magazine of Fantasy and Science Fiction that I got from Barnes & Noble. I am holding it as I sit in my car. The Apology Tour By Nnedi Okorafor HER LEFT EYE TWITCHED AND SHE wanted to slap the side of her face. But how would that look in front of all these people? She bristled at the thought and focused on the open book sitting on the podium before her. Her book. Her memoir. She'd lost her place. Shit, she thought. Ever since updating her glasses with progressive lenses, she'd been having this problem. They were supposed to have the opposite effect. She didn't think she'd ever get used to wearing glasses. Why would she? At sixty-four, she'd never worn glasses or contacts in her entire life, until the last year. She glanced up at her audience and softly laughed. She heard them all sigh, several people grinning back at her. Adoration. Interest. Despite it all. A big profile piece on 60 Minutes called her "The Stoic Unicorn." She had this. She'd had it for the past year, since her book was released directly into fanfare and instant bestseller status. And none of it had been surprising, considering who she was and what she'd written about. She was one of the most infamous people on the planet. Standing up here before two hundred people and reading from her book should have been the easy part. Her left eye twitched again and she grasped the sides of the podium. Then both of her eyes found her place and she continued reading. "Could I forgive myself, though? I created it. Then I perfected it. I was a top-tier engineer at Intelligent Robotics. I was a rarity in too many ways to list. I knew it. I was proud of it. I was on salary. I was doing... my job. What I'd been hired to do. But this project was my baby. It was my idea; it came out of my head. I am driven, ambitious, smart. Potential is worthless until realized into being. I am a doer. So I stayed up nights, researching, compiling, researching, digesting, thinking, theorizing, researching, testing and testing. I ignored my husband, my children, my grandchildren. I dropped out of their lives for those months. Ifocused razor.. blade sharp. Because I knew." She paused Why do I always read this part? But she knew that, too. It was what everyone wanted to hear. They would read her whole book. Yes. They wanted to know her full story. Yes. They were interested in her life, how she got here, to this place in time. Yes. They would do that on their own. Yes. However, right now, in this moment, on this book tour, they wanted to hear that hitch in her throat. See that frown on her brow. That pain. That guilt. Her left eye twitched, again. They couldn't see the twitching that had been plaguing her since the start of her book tour three months ago. If they could, they would enjoy, feel unburdened and refreshed by that, too. "I knew I could... wake it up. The singularity was right in my hands. I just needed to complete the alchemy in the coding. That's how I saw it. As alchemy. As I did it, I knew what I was doing. Tunnel vision did not limit my awareness of what it meant to create awareness. I knew. And..." She sighed and looked at her audience. They were waiting for it. It's therapy, she thought, not for the first time. Then she spoke the words, letting them tumble from her lips like heavy stones, "I'm sorry." She shut the book loudly, imagining a cloud of dust puffing up from its pages as if it were some ancient tome and she were the ghost of a sacrificed creature cursed long ago to read it forever. Stuck in the role of massaging the kinks out of the world's tense back. For creating something she couldn't have resisted creating the moment the path became clear to …
January 2, 2026 at 4:02 PM
Reposted by Fredy E
shawndoublea.bsky.social
SantaCorps: Snow Patrol

And the companion piece.

Seriously, I've had the pleasure of watching @ethanredd.bsky.social & @saredd99.bsky.social's work grow over the years.

Anything @virtuoso-neomedia.bsky.social is up to, you should follow.

store.steampowered.com/app/3398700/...
SantaCorps: Snow Patrol on Steam
Don the parka and shovel of the SantaCorps' Snow Patrol elf unit in this accessible arcade puzzle frenzy!
store.steampowered.com
January 1, 2026 at 1:31 AM
Reposted by Fredy E
susiewilliamson.bsky.social
Day 22 of Book Advent 📚
Nine Bar Blues by @shereereneethomas.bsky.social

Golden mermaids, curse-bound sisterhoods, poisoned landscapes & an otherworldly earth. Evocative, inventive & contemporary tales that haunt, uplift, & ascend, while moving with the rhythm of cultural change.
December 22, 2025 at 10:19 PM
Reposted by Fredy E
abeba.bsky.social
are you disgruntled by the current safety evaluation landscape? curious about what conceptual clarity, methodological soundness and rigour in AI evaluation might look like? if so, consider coming to dublin and doing a phd with me

apply here: aial.ie/hiring/phd-a...
About the PhD Audits and evaluation of AI systems — and the broader context that AI systems operate in — have become central to conceptualising, quantifying, measuring and understanding the operations, failures, limitations, underlying assumptions, and downstream societal implications of AI systems. Existing AI audit and evaluation efforts are fractured, done in a siloed and ad-hoc manner, and with little deliberation and reflection around conceptual rigour and methodological validity. This PhD is for a candidate that is passionate about exploring what a conceptually cogent, methodologically sound, and well-founded AI evaluation and safety research might look like. This requires grappling with questions such as: What does it mean to represent “ground truth” in proxies, synthetic data, or computational simulation? How do we reliably measure abstract and complex phenomena? What are the epistemological or methodological implications of quantification and measurement approaches we choose to employ? Particularly, what underlying presuppositions, values, or perspectives do they entail? How do we ensure the lived experiences of impacted communities play a critical role in the development and justification of measurement metrics and proxies? Through exploration of these questions, the candidate is expected to engage with core concepts in the philosophy of science, history of science, Black feminist epistemologies, and similar schools of thought to develop an in-depth understanding of existing practices with the aim of applying it to advance shared standards and best practice in AI evaluation. The candidate is expected to integrate empirical (for example, through analysis or evaluation of existing benchmarks) or practical (for example, by executing evaluation of AI systems) components into the overall work.
December 17, 2025 at 7:33 PM
Reposted by Fredy E
bunnie.org
Hey! I'll be giving a short talk about IRIS tomorrow at the @hackadayofficial.bsky.social Berlin event at noon. I also have a couple copies of The New Essential Guide to Shenzhen to give away to anyone who wants it. Come by and say hi if you happen to be attending the event!
March 14, 2025 at 11:05 AM
Reposted by Fredy E
bunnie.org
In case you missed it - my talk on non destructive inspection of silicon at 38c3 is now up: media.ccc.de/v/38c3-iris-...

Thanks to everyone who stopped by and said hi at the congress!
IRIS: Non-Destructive Inspection of Silicon
IRIS (Infra-Red, *in situ*) is a technique for non-destructively inspecting the construction of a select but common type of chip. It can ...
media.ccc.de
January 3, 2025 at 7:39 AM
Reposted by Fredy E
susanrinkunas.com
Thank you Rose!! Link for others interested:
buffer.com/bluesky
Schedule your Bluesky posts with Buffer | Build an Authentic Community
Buffer brings together your Bluesky and other decentralized platforms like Mastodon and Threads, making it easy to manage your content and community from one place. Try Buffer free for 14 days or enjo...
buffer.com
November 26, 2024 at 4:20 PM
xfredye.bsky.social
I found the articles wrote by @khari.bsky.social very ilustrative about the actual state of AI regulation in United States.

Here is a good article for understanding at what point is regulation today in USA
calmatters.org/economy/tech... #AI #regulation #Algorithms
What California lawmakers did to regulate artificial intelligence
More than a dozen artificial intelligence bills are headed to Gov. Gavin Newsom’s desk to either get vetoed or get signed into law.
calmatters.org
November 26, 2024 at 3:30 PM
xfredye.bsky.social
Changes are there every day, at every scale of nature. Opposing to them is a stupid move.

It is better to be part of those who lead and control de changes, instead of being part of the resistance.
November 21, 2024 at 6:12 PM