

Ew. Reviewing code is so 2024


Ew. Reviewing code is so 2024


This is one of those things that break two golden rules. Minimal change, or consistency? The only further variable a most senior person considers is the brittleness of the environment and deployment process.


Bravo. An LLM is an amazing tool in software dev.


It’s an Israeli and US company. If anyone is somehow MITM snooping your TLS traffic, it’s them.


So is that a yes? AI slop?


This is amazing


This is what I was gonna mention. Even with a 24gb 4090 I have no context room. I run it 8-bit quant on a DGX Spark for that. It’s only 13 tps but 256k context.
I’m native English speaking and it sounded fine. I didn’t recognize any “style” that would be stereotypical of anything. It read like an engineering blog post.
As a long time financial supporter, thank you! Is this your full time gig yet?
they don’t have ai…


I hear ya. I’m 50 years old. Super safe driver, also never been in an accident. It’s wild that you had such a bad experience, but I believe you.


What year Tesla? Sounds like my 2023 (HW3). The follow closely thing seems to have gotten better in the last couple updates tho.


64GB unified memory. I run it (and a lot more) on a dgx spark, but a Mac mini would suffice also.
You could prob run 4-bit version on a RTX card with 32g. Maybe even 24g. Like a 5090 or 4090 or such.
So much info out there.


I run Qwen 3.6 27B at home. For “free”. It is extremely useful.
My point being that I’m not going to be priced out of using it
Docker does by default - it only works if you use sudo. But the docs tell you to add yourself to the docker group (which requires sudo to do). Then running docker doesn’t require sudo anymore.


That seems similar to my path. Covid put me on the remote worker world, so we packed up and hit the road. Moved to a remote small town near vast nature. Bought a nice house on the river.
I’m also self employed and doing normal near retirement things. Currently on a 3-6 month contract for an ex employer in the financial space doing AI security stuff.
Then I’ll take the winter off and snowboard every day.


I’m 50 and in the process of retiring. It’s such a fitting exit. I’ll ride the slop wagon into the sunset.
Qwen 3.6 27B dense is really good. Very usable coding output
Yeah, would be best within particular client implementations. Not core Lemmy tho. Too many ways to do this and high maintenance.