lnenad

I actually made something that has that exact purpose, making reviewing easier. https://github.com/lnenad/difiko

I actually made the same thing in Rust https://github.com/lnenad/difiko

Just FYI a large portion is vibecoded for those that don't like that.

Cool demo, but even the smallest model does 2.5 tps on my 7900xtx which is not great.

Another happy user here. Thank you for the awesome project.

I don't hate AI companies. The key value proposition is gather data > feed it to AI for semantic analysis (does the actual work, is a prompt) > display it in a UI

It's still a prompt, it's just not a static one. Either way props for building a company from it.

I thought startups wrapping prompts would require something a more complex than semantic analysis, which is literally what this is. And for 500 bucks. Wow. Props for being able to sell this.

I don't get the appeal of the UI, why is it so complex/convoluted.

But 10% is 50 million square kilometers. If we project this surface area to a theoretical planet with a diameter of 2000kms this planet alone would support 2000+ Starlink satellites in LEO. This is again not negligible.

Yeah, nothing exists in a vacuum :), even for concrete columns rebar is added.

That's true, but it's still factor and if price comes down having that much compressive strength at your disposal could be an interesting choice for supertall skyscrapers.

Yes, I agree! But a 10% increase in surface area of LEO isn't small when compared to the size of the satellite, taking into account how the surface area at 10km compares to airplanes.

Respectfully disagree as you're comparing the surface to the size of the object, so it definitely matters.

Here's some math:

Average Earth diameter: 12742kms + 10km Average airplane surface area = 500m2 12752^2*pi = 510,865,389km2 Surface flight/plane = 1021730 planes

Starlink orbit height = ~500km Surface at orbit = ~551,712,377 so ~8% increase (which is non-negligible) Average Starlink satellite surface area = ~7m2 Surface LEO/satelite = 78816053 satellites (77x compared to airplanes)

Daily flights 50k-100k. Total number of satellites <20k.

And this is only for Starlink LEO. If you go for higher orbits the surface grows substantially. Also satellites have predictable paths, altitudes, airplanes maneuver and turn, gain altitude/lose altitude. They gather around points (airfields) etc...

High compressive strengths are not very useful.

I mean, it depends on the use case right? Modern tall buildings/skyscrapers with metal framing use metal pillars where high compressive strength is very useful.

Surface of a sphere (spheroid) is the square of the diameter. Planes fly at ~10kms, satellites at orders of magnitude higher.

I'm building a low friction logging/metrics platform that allows for easy visibility into your apps. The idea was to make something that isn't really geared towards enterprise and can be used by anyone. I started about 6 months ago and have been testing extensively so far with an internal beta with a few colleagues/friends.

https://logdot.io

Considering the DeepSWE result (imho if you're gonna give value to benchmarks this is one of the best) it's not good enough.

I get it, you just prefer to do things differently

Faster iteration means i mentally checkout less and am more involved with the code being created.

This is a good point I didn't consider and you're right. More interaction brings you closer to the code.

I still think that this is the opposite of what I personally want. Either I write the code (or a large majority of it), and be fully involved; or be more disconnected but more free to focus on other things. The middle ground removes me from the equation, but also requires me to babysit.

You can ask for a complete feature/app/business. Or you can split up the work into verifiable/testable pieces and rely on a high quality AI to deliver. As time goes by the pieces will get larger as capability grows. I still trust myself and my experience when arch is involved, but AI has been great at tackling lower level stuff. And with Fable I don't really care it takes a while for it to complete, as I know I can trust it a lot more (which is what I personally prefer). Yes, with a 10k tps model you can iterate quickly. But that's not me personally.

Again, my point is exactly the opposite. Higher quality implies a mistake isn't made in a significant % of cases.

Why? I'm personally on the opposite end. Less babysitting/higher quality means more time goes back to me/the user. 1000tps of bad code means you have to keep validating the output and circling back.

For me it's the skill. So many things you have to be good at to build this. Really impressive and slightly jealous.

Very nice! Thank you for building this.

What wild regurgitation of some generic sentence is this lol?

If it's my data/money getting stolen, I'd give no fucks about MSI getting a fine or whatever the usual reaction to these fuckups is. On the other hand, if I found an exploit and there wasn't a bounty available, I'd still report it. Betterment of the world and all that.

But it's not "MSI computers" it's actual people getting pwned.

Ok? I agree with everything. What does that have to do with reporting exploits that don't have bounties?

You're actually helping the people that use the software from getting pwned, companies are secondary beneficiaries.

You understand the concept of doing something that doesn't bring direct monetary benefit?

Did I ever say any of those things?

I'm not justifying anything nor am I arguing against it. I'm saying it's not a thing worth mentioning as it's not related to AI in any way. If one would try to find AI related negatives go for datacentres, chip shortage/price hikes etc... So many better arguments to choose than just saying "CEO's are lying!!1!".