fathermarz

The website on mobile is constantly jumping around so I literally can’t read the copy before the height changes and goes below the fold.

Also, use cases? I’m a little lost after nailing down all the movement.

> Service Misuse. You acknowledge that without the written consent of us and/or the relevant rights holders, (i)you have no authority to use Kimi and the content generated by Kimi in any commercial manner; (ii)you may not use our Services to develop products or services that compete with us.

https://www.kimi.com/user/agreement/modelUse

Terms of use are very broad and not friendly for most things.

Can’t use for commercial purposes. Can’t opt out of training. Data retained.

The issue with software supply chain is that you have to be in step the entire chain and developers are creative. I don’t have to pull in a dependency when I can copy paste code out of third party repos or stack overflow.

Also this doesn’t take into consideration one of the ugliest sides of SSC which is M&A. Sure on a completely greenfield piece of software, seems like it would be a solution (although over engineered tbh). But that doesn’t really exist in closed source systems as they are Frankenstein pieces of software that get new stuff bolted onto them.

Just got mine. $534,366,582,647.75

Building Critical Infrastructure protection software that helps security teams create real grounded processes that don’t live in spreadsheets and slide decks. https://cabreza.com

Most critical infrastructure orgs don’t have the budget to hire consultants, and even if they do, the deliverable is a deck, or a spreadsheet, or a PDF. We want to help any org of any size create a security regimen outside of these stale and disparate docs. For FREE.

Plus we have additional tools that we are building on top of the free software that will help in other areas besides policies and procedures. Like OSINT of any orgs operational and physical footprints.

Knowing both doctors and nurses before AI scribes, actually hand writing notes isn’t usually done in front of the patient. They make quick notes so they keep up with the conversation and then when they need to do long form, they reference their scribbles but do the rest from memory at the END of the day.

For every one person that says they mind, it adds likely an additional margin of error AND time to their day. So I really disagree with making people’s lives harder. That’s why technology exists…

And more and more of these systems are on device models, not cloud compute. So I don’t see that argument either

Thank you for the education. What is the consequence of an offence?

Revenue generating services aren’t the same as critical infrastructure. This bug, I would argue, does not hurt creator’s revenue in a substantial enough way to call it “safety”.

I think there is a fine line. YouTube is not critical software and no one’s life depends on the safety (putting mental health aside) of the code running. Some software engineers do however write code that is critical, but to your point, I don’t think they are ever considered liable.

I went through an acquisition as a Canadian software developer getting acquired by an American company. They wanted us to be called engineers like the rest of their SWEs but in Canada it’s a protected namespace. It’s illegal to call yourself an engineer without having the ring and the papers. Which personally I can appreciate.

I’m not sure I understand. Babysitting models is not a multiplier IMO. If you have done 1000s of turns your harness should get sharper and less likely to go off the rails.

Also I find that on greenfield, babysitting is a must, but once you have established your house style of patterns, abstractions, and baselines, you can let any of them roam free cause they will look for examples before going forward.

I agree with the sentiment though that if you let a swarm design and code your whole codebase, you will be lost in how it fits together. More feature bloat than code bloat though from my experience

Good read and good breakdown. I feel like this is where I am in my journey is letting that roll off.

Second this

This is what life is about right here

When ChatGPT 3.5 dropped, I typed in. “Can you make me a Python script that tells me when the ISS passes over my house at <address>” and it one shotted it. I knew the world had changed forever.

I sincerely disagree that AI is worse than the crypto/NFT hype… pig butchering is one of the most disgusting practices imaginable and it was turned into a legitimate low effort vehicle for scammers due to web3 and the hype train.

AI is definitely on a scale of magnitude more but it has inherent value outside of “scarcity”. It’s actually quite the opposite with sheer supply/demand balance. Also investing in crypto made me less money than investing in myself by using AI to learn and challenge myself to think differently.

Excellent write up and inspired me for our next IA design run. After reading Fly’s Litestream work it makes me think this is a solid option.

I’m confused, isn’t this rewrite still unreleased as of today? Surely people understand that a simple, “do an audit for memory safety” will bring it up to par.

100% could not agree more. Was hired to fix a vibe coded app for a company that one of their people whipped up over five months. Tens of thousands of lines of code in a single file and no security whatsoever.

The guy uses Claude Code, same as me… it’s like a highly skilled mason with a chisel, against me with a chisel. I’m not going to produce the same masterpiece, because there is SME that underpins the accelerant.

The slot machine lever is my least favourite opinion on the subject.

Also, let’s not forget. The developer is rarely the person pitching the feature, and is normally given the constraints and the PRD…

Soooo people can keep tiptapping on the keyboard, but eventually they need to open their mind to the possibility that “the old way” is actually dead.

Completely agree. This is a harness problem, not a model problem. The model is rarely the issue these days

I built a community tool for exactly this, based on privacy first principals but around the what. It’s workflow based and not “put your sensitive data into ChatGPT and hope it captures the right stuff”. Mostly built for security folks but anyone can use it

Check it out: https://redact.cabreza.com

I have been seeing this messaging everywhere and I have not noticed this. I have had the inverse with 4.7 over 4.6.

I think people aren’t reading the system cards when they come out. They explicitly explain your workflow needs to change. They added more levels of effort and I see no mention of that in this post.

Did y’all forget Opus 4? That was not that long ago that Claude was essentially unusable then. We are peak wizardry right now and no one is talking positively. It’s all doom and gloom around here these days.

Couldn’t agree more with this sentiment. Bootstrapping leaning AI-first means I can iterate a few arch designs and actually test it, stress it, and rip it out if needed, within a few hours.

The technical debt problem is real, but as long as after a large session across 10-12 repos over a couple days, I can do a sweep for loose ends and kill dead code that we had for an old implementation. It’s less about building a piece, and more like building version 1 of a feature and then building version 2 a week later instead of 6 months later.

Building Cabreza Command (https://cabreza.com/product). Most critical infrastructure orgs manage their OT security program across SharePoint folders, Excel trackers, and slide decks that get updated once a year. The actual state of the program lives in someone’s head.

Command replaces that with a platform that maps to their real sites, real assets, and real operational constraints, so they can actually run the program, not just document it.

Consulting firms use it to deliver more engagements with the same team. Asset owners use it to keep the program alive between engagements, or run one themselves.

I love this and it’s very clean. Only piece of feedback, put the play again button where the scores are as that is where the thumb sits on the phone to do the tapping. Mind numbingly groovy

I am not sure about the “build it yourself is expensive” argument. If you were non-technical, you probably shouldn’t try to build it yourself, but for technical folks, that’s not a stretch.

I have seen this when approaching ~30% context window remaining.

There was a big bug in the Voice MCP I was using that it would just talk to itself back and forth too.

I didn’t understand the hype on Medvi in the first place. I don’t think this type of credit can be given to “two people” when it actually makes money by being a middle-man service, selling other peoples skills, and other company’s products.

I highly doubt they used AI to build the network…

This is a skill in life with people as much as it is with LLMs. One should always question everything and build strongman arguments for one’s self. Using a pros and cons approach brings it back to reality in most cases, especially when it comes to _serious matters_.

It’s less about “challenge my thinking” and more about playing it out in long tail scenarios, thought exercises, mental models, and devils advocate.