Nostr View

Xtr3m3hodl
@Xtr3m3hodl

230 following 5 followers

Xtr3m3hodl @Xtr3m3hodl - 2d

Don't you think it depends on the LLM provider? And when you say better, it isn't obvious what you mean since one would imagine the types on typescript would act as a feedback loop to the LLM when something is off?

1
0
3

Xtr3m3hodl @Xtr3m3hodl - 3d

😂👌 beautiful

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 3d

Torrenting is making a big comeback. Especially in the area of TV subscription 😅

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 5d

Just saw this in my mail as well. Checked the sender id and found out it wasn't from trezor.

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 6d

Npubs are better phone numbers than actual phone numbers. I like that I can assign custom usernames to npubs in #keychat

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 6d

Any Theo Von's podcast episode. Just fun listening to him.

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 6d

Even when working with others, and especially on existing projects, that's how I mostly work. I'm still familiar with all the obvious software development practices. It is just, I don't dwell too much on a written well thought out plan (things change more often). The way I see it, the "plan" is being developed mentally as I attempt to solve the problem

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 6d

I don't know if it's just me but I dislike having to come up with a comprehensive plan of what I need to build before build it. I would end up getting tired and loosing interest completely unlike just jumping on the problem and going with the flow, backtracking when new information comes up to ensure things are done well and actually testing hypothesis on the fly. It sounds counterintuitive to how most software engineering processes work but it hasn't failed me yet for the 15+ years building software. Anyone else experience this #asknostr

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 7d

Another banger from PewDiePie https://youtu.be/u_Lxkt50xOg

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 7d

I was listening to a podcast where the host was advocating rewrites when using LLMs in order to avoid technical debt.

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 7d

Anyone that preaches the benefits of always re-writing code from scratch when it gets complex with LLMs obviously never works with backend systems. Because if they did, they would realize how absurd that is.

0
0
2

Xtr3m3hodl @Xtr3m3hodl - 8d

https://youtu.be/maiezdDneek

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 8d

This would be a great feature if eventually supported

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 9d

I can understand an AI based IDE but I'm drawing a line at an AI terminal. I don't get the appeal for the warp terminal.

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 9d

An interesting read about anthropic controlled experiments on agents blackmail https://www.anthropic.com/research/agentic-misalignment

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 9d

This is just beautiful. Well done 👍

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 12d

YouTube seems to have fixed their algorithms . This is just purely hilarious 😂 https://youtu.be/nmrOeS-r7P0

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 12d

Never paid any attention to it

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 12d

The best part of this talk IMO was about building more augmented systems and less autonomous systems

0
0
3

Xtr3m3hodl @Xtr3m3hodl - 12d

Just pieced that now. Can't believe I didn't think about that sooner

0
0
3
Showing page 1 of 11 pages