lysecret

6 ms·

61.
Maestro 2.0(maestro.dev)
1 points by lysecret 1y ago|0 comments
62.
by lysecret 1y ago
I love ruff and I love uv and I’m sure I will ty as soon as it’s ready but please keep them separate.
63.
by lysecret 1y ago
Oh god what is this website it gives me a headache with all the pop-ups and auto playing videos.
64.
MCP vs. CLI(mariozechner.at)
4 points by lysecret 1y ago|1 comments
65.
by lysecret 1y ago
The designing data intensive Applications we actually needed (nothing against the original but this one is definitely more practical)
66.
by lysecret 1y ago
Bit of a click baity title since thy can definitely help in building software. However, I agree with the main thesis (that they can’t do it on their own). Also related to this this whole idea of “we will easily fix memory next” will turn ou
67.
by lysecret 1y ago
Just wow. My greatest respect! Also an incredible write up. I like the take-away that an essential ingredient to a search engine is curated and well filtered data (garbage in garbage out) I feel like this has been a big learning of the LLM
68.
by lysecret 1y ago
Are you using ai for the parsing/setting up the parsing structure?
69.
by lysecret 1y ago
Couldn’t believe it was real haha
70.
by lysecret 1y ago
I honestly think the times where you became employable simply with decent grades and a degree are over.
71.
by lysecret 1y ago
It’s so fascinating that we just keep on finding more positive effects.
72.
by lysecret 1y ago
Fully agree, from the original post: "and forget that the code even exists."
73.
by lysecret 1y ago
Yep this is pretty much what I do. I also added a /check command to explcicitly runn all those checks + pyright and ruff format.
74.
by lysecret 1y ago
Yep, I prefer pyright though, but ty is too early to be relied on (though I love love ruff so I'm sure they will get there.(
75.
by lysecret 1y ago
I have been coding with claude code for about 3 weeks and I love it. I have bout 10yoe and mostly do Python ML / Data Eng. Here are a few reasons: 1. It takes away the pain of starting. I have no barrier to writing text but there is a
76.
by lysecret 1y ago
The core thesis is that your types received by the api should not be the same as the types you process internally. I can see a situation where this makes sense and a situation where this senselessly duplicates everything. The blog post show
77.
by lysecret 1y ago
IMO Claude code was a huge step up. We have a large and well structured python code base revolving mostly around large and complicated adapter pattern Claude is almost fully capable to implement a new adapter if given the right prompt/
78.
by lysecret 1y ago
Main reason is (used to be) performance. But since pydantic 2.0 it’s not an issue anymore I default to use pydantic for everything.
79.
by lysecret 1y ago
After what we have seen in terms of talent acquisition this seems like a no-brainer really.
80.
by lysecret 1y ago
Excellent commentary from Kaparthy on RL: https://x.com/karpathy/status/1944435412489171119
81.
by lysecret 1y ago
I'm pretty sure we will, because we already have? Remember work used to be 6 days a week.
82.
by lysecret 1y ago
This feels very Ai generated.
83.
by lysecret 1y ago
I'm doing the exact same thing. I have built a small app for it too but I like using the plain ai too.
84.
by lysecret 1y ago
Good moment to reread the grand inquisitor to be prepared.
85.
by lysecret 1y ago
So what's its going to be in the end Diffusion or Auto-Regression? After OpenAi (probably) released an Auto-Regressive model for their image generator I thought things might sway the other way.
86.
by lysecret 1y ago
Can we all just take a step back and appreciate that this is a post mortem on a personality bug of an AI that hundreds of millions of people use every day. What a world.
87.
by lysecret 1y ago
Funny you call it value I call it inefficiency.
88.
by lysecret 1y ago
Ye its also funny to me. On the one side people are saying: Look we have computer use, browser use etc. so we don't need an api! And on the other side saying, look apis are way too complicated we need our own protocol!
89.
by lysecret 2y ago
You can try out everything for 300 dollars easily. Most expensive thing you can do is get a server with 8 h200s and spend 90 dollars an hour.
90.
by lysecret 2y ago
Never seen so much positivity related to a tool since SQLite

More ›