The unadulterated racism expressed in this video by both Brandon Gill and Benny Johnson is breathtaking in its alignment to Nazi propaganda of the 1930s.
MAGA -> Haitians & Somalis
Nazis -> Jews
Autonomous platforms feature in various episodes of @practicalai.bsky.social. Sometimes China comes up and sometimes it doesn’t.
Economically asymmetrical warfare >>>
Expensive exquisite systems are being taken down by off-the-shelf commodity systems.
Just wait till drones are actually swarming…
A masterclass in presidential negotiation strategy!
And the earth was without norm, and void;
We’re looking forward to learning more about the GUARDRAILS Act on @practicalai.bsky.social
Episode 349:
Humility in the Age of Agentic Coding
---
In this episode we talk with Steve Klabnik, known for his work on the Rust programming language, about his firsthand experience building Rue using tools like Claude and what it reveals about the future of coding.
#AI
practicalai.fm/349
Episode 348:
AI policy and the battle for computing power
---
In this episode, Ben Buchanan, Assistant Professor at The Johns Hopkins University and former White House Special Advisor for AI, joins us to discuss #AIpolicy, importance of computing power, and AI governance.
#AI
practicalai.fm/348
www.youtube.com/watch?v=DfmV...
@rustcodepro.bsky.social and @filtra.bsky.social -> you might be interested in the interview we did with Nathaniel Simard, creator of Burn, on @practicalai.bsky.social
practicalai.fm/242
Generative AI isn’t good for _everything_.
It’s good for some things, but not for others.
It’s one very powerful tool in an AI toolbox that includes an array of tools.
That’s what we share with our audience on the Practical AI podcast. We don’t hype generative AI. We have reasonable expectations.
I’m an AI research engineer and cohost of the Practical AI podcast.
I think you aren’t approaching generative models with realistic expectations. Use them for what they’re good for, but go to non-generative models where generative models aren’t strong. Generative models are good for some tasks.
Yep
I’m the cohost of the @practicalai.bsky.social podcast and have often talked about Python’s stranglehold on AI, and the hope that Rust can make inroads on that.
It’s not just the Python ecosystem, but that historically Python has been the go-to language of the larger scientific community.
I had a great time being interviewed about fully-autonomous swarming and other topics on this joint @changelog.com / @practicalai.bsky.social episode!
Neither. You’re offering two bad choices.
Per Microsoft CTO Mark Russinovich, C/C++ should be deprecated and never used for new projects.
A “Rust programmer” too lazy or incompetent to learn the borrow checker should consider a career change or grow up. The problem is the person, not the tool.
That’s not “mastering the borrow checker”. It’s avoiding the borrow checker. Better to understand its rules and design your code to be compatible with them. Using indices is a data-oriented design pattern for specific architectural challenges, not a strategy to bypass the borrow checker entirely.
Rust has value without the borrow checker, but removing it is shortsighted. Learning ownership to master the borrow checker makes one a better programmer. Once learned, velocity is no longer impaired and Rust is easier to work and prototype in since you're no longer fighting with the borrow checker.
I love Go!
I spent a decade coding in Go. It’s great for systems programming so long as you don’t need blazing speed or to be very low level - and the garbage collector pauses are acceptable.
I ultimately needed blazing speed, very low level, and no garbage collection.
So I switched to Rust.
🤷♂️🦀
Are you using a tutorial like one of these - or something similar?
• Writing an OS in Rust: Philipp Oppermann's blog
• The Learnix Operating System
This is a beautiful true Christmas story. Take a few minutes to read it and imagine a world in which we all exhibit similar love and generosity to our neighbors - and to the animals around us.
Merry Christmas from our family to yours. 🎄🥰
www.bbc.com/news/article...
The First Amendment protects our right to record law enforcement activity in public.
Thanks for the post! Do you have any opinion regarding Claude versus ChatGPT (or others) specifically for quality, accuracy, precision, etc., in generating compilable Rust code? They all do Python moderately well, but in my experience less so with compilable Rust code.
Today’s speech by Blaise Metreweli, Chief of the United Kingdom’s Secret Intelligence Service (MI6).
This is the speech that I wish I was hearing from U.S. intelligence leaders.
www.gov.uk/government/s...
There is a lot of complexity - both algorithmically and in terms of practical engineering - to effect true swarming.
Boids only addresses flocking behavior, which in this context is a subset of swarming.
There are many considerations to solve and integrate.
And there are many types of swarming.
Truths that make us uncomfortable.
replacement.ai
All good! Thank you for listening! We’re at a weird moment with swarming. It’s a giant buzzword, but nobody has actually done it yet. What they are calling swarms are “not a swarm”. It’s a good moment to jump in and ramp up on the possibilities. Good luck!
Have you considered detailing your workflows in a blog post?
Speaking as both a Rustacean and a licensed wildlife rehabilitator who specializes in raccoons, I think you should ensure this non-traditional student gets a first-class education in Rust. Raccoons are extremely intelligent. Cheer it! 🦀🦝
Uh oh, I have no idea what I said, or what it was in reference to. And when I said it I doubt I realized who you were, though I do now. Please forgive me.
