⚡Hữu Lũng 🧗🌄🌾🇻🇳

8.2K posts

⚡Hữu Lũng 🧗🌄🌾🇻🇳 banner
⚡Hữu Lũng 🧗🌄🌾🇻🇳

⚡Hữu Lũng 🧗🌄🌾🇻🇳

@thlorenz

Building at @magicblock.

Katılım Ağustos 2010
201 Takip Edilen3.2K Takipçiler
⚡Hữu Lũng 🧗🌄🌾🇻🇳
When you read old scifi stories and they make video calls connecting to the other person via a rotary dial 🙃 📺☎️
English
0
0
1
119
⚡Hữu Lũng 🧗🌄🌾🇻🇳
So the codex installer overwrote my entire bash profile. Fortunately a simple prompt with @AmpCode and the fact that I got my dotfiles source controlled allowed a quick fix. Welcome to the age of AI!
English
0
0
5
507
⚡Hữu Lũng 🧗🌄🌾🇻🇳 retweetledi
Felix Geisendörfer
Felix Geisendörfer@felixge·
In exchange for 18 minutes of your time, you can get an elevator pitch about why you should be using pi (instead of claude) and perhaps more importantly, one of the sanest takes on using AI effectively 🔥 youtube.com/watch?v=RjfbvD…
YouTube video
YouTube
English
0
3
8
1.2K
⚡Hữu Lũng 🧗🌄🌾🇻🇳
OK so @github runners seem less beefy and are a lot slower than before. @OpenAI's codex uses more credits for same workloads with same model and I run out much faster than before. What a time to be alive. NOT
English
0
0
0
166
⚡Hữu Lũng 🧗🌄🌾🇻🇳
> Me: Did you seriously just ... ? > Codex: Correct. I missed your explicit instruction and should not have done that. 🤦
English
0
0
0
137
⚡Hữu Lũng 🧗🌄🌾🇻🇳
Agents are becoming more human like. When asked why it removed useful comments, codex responded: "Those comments disappeared because I rewrote proto/event.proto" Sure they "disappeared". Learning to avoid taking responsibility and come up with excuses instead 😂
English
1
0
1
113
⚡Hữu Lũng 🧗🌄🌾🇻🇳
This is what "coding" looks like in 2026. Step by step narrowing down each feature/problem until I have a detailed plan I agree with. Then implement step by step. Very little "direct" coding on my part.
⚡Hữu Lũng 🧗🌄🌾🇻🇳 tweet media
English
0
0
4
194
⚡Hữu Lũng 🧗🌄🌾🇻🇳
Please codex I really couldn't care less why you do something or every little blip that's going on to lead to a decision. Just do the thing and keep the rest to yourself 😆
English
0
0
0
153
Noah
Noah@redacted_noah·
@thlorenz @jacobvcreech Sadly it doesn’t work if you used squads and buffer deploys to upgrade your IDLs
English
1
0
1
151
Noah
Noah@redacted_noah·
Hey @jacobvcreech foundation should run an indexer that watches the metadata program an indexes the history of each IDL, including all versions. It should then expose a service that (1) gives me the current IDL for a program given a major version (so two majors can be active at the same time) and (2) gives me a history of every IDL change for a program between two timestamps, so I can make changes as I index. With Streamingfast + Claude you could probably vibecode this in a day.
karsten@karstenalytics

@redacted_noah the IDL situation is indeed unsatisfying… I started to build an IDL history for the projects I cover myself to solve this. Can still work nicely if you run it automated and potentially even let Claude create issues/ PRs if an IDL change would break something (on my side)

English
6
1
40
5.2K
Jonas Hahn
Jonas Hahn@SolPlay_jonas·
This sounds like a good idea! Did someone build this already maybe? Otherwise we could maybe build it into the explorer and expose as api. 🤔
Noah@redacted_noah

Hey @jacobvcreech foundation should run an indexer that watches the metadata program an indexes the history of each IDL, including all versions. It should then expose a service that (1) gives me the current IDL for a program given a major version (so two majors can be active at the same time) and (2) gives me a history of every IDL change for a program between two timestamps, so I can make changes as I index. With Streamingfast + Claude you could probably vibecode this in a day.

English
4
0
10
1.3K
⚡Hữu Lũng 🧗🌄🌾🇻🇳
When asked to generate tests for a specific module I found agents overly eager to please omit tests that show flaws in my code. Obviously not what you'd want, so manual review is required or maybe write the tests first and/or indicate that they should ensure specs, not just pass.
English
1
0
1
235
⚡Hữu Lũng 🧗🌄🌾🇻🇳
I'm not convinced that @AmpCode smart mode is considerably better than rush. It even seems to make more mistakes sometimes since it tries to take larger steps, while rush that goes piecemeal and then fixes issues one by one.
English
1
0
1
260