Update; this is now the first AI Agent native QRNG! MCP server ready to go. 🦞🔗⤵️

Do I need a Dell Pro Max GB10? Absolutely not; do I want it? Heck yeah 😅

Completely agreed! The only sustainable approach is structural: isolate workloads, reduce lateral movement, and make it so a compromised model or agent can’t take the whole system down. Treating model execution environments more like hazardous material labs than open workspaces.

A photo of James and I speaking to a room of folks at Rejekts about Type-2 Hypervisors

Enjoyed doing our talk today on hypervisors at @rejekts.io!

Was a blast! As Rejekts always is!

Infrastructure Security Lounge @NVIDIA GTC · Luma Drop by the Infrastructure Security Lounge at NVIDIA GTC to connect with fellow engineers, technical leaders, and founders working on distributed systems, AI…

Heading to NVIDIA GTC in San Jose? Already tired and you haven't even packed? Plan some relaxation into your schedule with the Infrastructure Security Lounge. Escape the conference chaos, enjoy some snacks and drinks, and follow the conference livestream. Register in advance, or just pop in!

Edera X Google Cloud Marketplace

Edera is now live on Google Cloud Marketplace.
You can procure Edera directly in GCP — and use your existing cloud credits to secure your container and AI workloads.
Full isolation. Near-native performance. No procurement friction.
Security without compromise, now one click away.
edera.link/gcp

Quick update here local models are not the way to go for this tool, mainly because of system prompt size

When does a threat actor engage in a 💎🤲 position with this new framework? 🤣

So far I'm playing with LFM2.5-1.2B by Liquid AI. Thankfully I didn't wire in Claude because accounts are getting banned for it.

Bonus: I’m running local inference for the tool directly on the Mac Mini using LM Studio. For compute per dollar, that’s honestly some of the best value out there.

The serious answer though: some workflows really do benefit from a device like this especially when you’re trying to avoid website rate limits or need to integrate with actual macOS apps. It’s not quite the same comparison as a Raspberry Pi for that reason.

Okay, I’ll admit it I just needed an excuse to finally buy a Mac Mini (been wanting one for way too long).

P.S: The Clawd/MoltBot product has in-depth documentation and tooling to avoid security issues. I'm sure as an awesome open source project it will evolve overtime to be more secure by default.

Expecting people to "not use" technology because it's unsafe is a losing proposition for consumer or enterprise apps. It's our responsibility as people building in technology that products are secure. Passing the blame to the user is 1) not an excuse 2) a sure way to not succeed as a business.

For the paranoid a hardware security key where applicable is even better. (I'm looking at you U.S Banks with dismal security that still don't support hardware keys for customers and rely on SMS).

b) don't enable 2FA on everything (whether authenticator code apps or passkey vs SMS based 2FA) you can consider yourself lucky that your accounts have not yet been compromised.

Forget about how irresponsible your Clawd/MoltBot usage is if you don't a)use a password managers for unique passwords per account

Prompt injections are real and there is no real solution for them... yet.

I barely stepped away from “geeking out” for less than 48 hrs and apparently I’m supposed to be running Clawdbot and should have purchased a Mac Mini for it… Don’t tell my wife but I already did. Also I’m all up for being in tech to constantly learn but can we slow down a bit 🤣

Does anyone else try to completely maximize their Claude Opus weekly allowance before it resets? It’s like going to a fancy buffet and repeating the fanciest dish one last time right?

I have been using Perplexity Pro for about two months and I haven’t once since then used Google does anyone else feel like traditional search even Google’s AI assisted search is severely lagging behind?

The A in AGI stands for Ads Had to crunch the ad revenue projection numbers for OpenAI real quick, had enough of the fearmongering and nonsense takes. AGI... now funded by 5x60 second unskippable ads!

Ha! ossa-ma.github.io/blog/openads

Daniel Fernàndez (@dnlfdz) AI Assisted White Collar Work: Optimistically Pessimistic A while back, we heard about tools limiting usage because of abuse in coding tasks. At the time, I couldn’t really grasp how someone could hi...

Post test limits here are annoying me a bit so giving Substack a shot. substack.com/@dnlfdz/note...

How long until Enterprise Software / Productivity suites cutoff API access to competing AI agents? I’m going to guess we end up with walled gardens. Do companies like Box/Dropbox have a MOAT for “holding the data”? We shall see.

Well, it’s going to be a weekend full of fun tests with my new and flashy “WiFi 7 Router”

Those are still some impressive numbers.

Please don’t tell me you rocking just a t-shirt.

Economics Rant: has anyone considered the premise that maybe gold price is not as high as we think and maybe the USD is just worth exponentially less overtime? Anyway; back to printing money.

Product Futurists - How Product Management Must Evolve in the Generative AI Era A few weeks ago, I watched something that stopped me cold. In a modern developer workflow, multiple AI agents were running code implementation in parallel; planning, executing, and iterating faster th...

Product Management is evolving fast. Join me and other product professionals exploring how product leaders can thrive in the age of intelligent systems. Let’s shape the future of our discipline together.

www.linkedin.com/pulse/produc...