andyrewlee
1.1K posts
andyrewlee
@andyrewlee
coding agent connoisseur
San Francisco, CA Katılım Nisan 2009
262 Takip Edilen285 Takipçiler
@josevalim i combined the concept of a branch and worktree and it's been working well for me.
worktree with the same name as the branch gets created and deleted with the branch.
once you set up scripts to automate, it makes worktrees invisible since they are just branches.
English
What is your favorite git worktree integration for coding agents? Why?
English
Been so much fun cooking OpenShell and NemoClaw with the @NVIDIAAI folks! 🙏🦞
Huge step towards secure agents you can trust.
What’s your OpenClaw strategy?
English
Agent/AI literacy will be a primary school subject in the next 3-5 years
How to use and work with agents is going to supersede most other subjects in importance
Similarly, robot literacy will follow in 5-15 years
English
@Jeff_Burke14 i think people are reacting to the first statement in the post. opposite of jumping into the details.
no one is saying AI (even without replit) can easily build prototypes, internal tools, etc. no need to vilify developers.
English
The comments are what you’d expect. Mostly developers angry and jumping into the details, even accusing us of malice.
For many prototypes, internal tools, SMBs, etc, the developer is not necessary. Sorry if you don’t like that.
More to build, but the trend line is clear.
a16z@a16z
"Not having a coding experience is becoming an advantage." Replit CEO Amjad Masad: "You don't need any development experience. You need grit. You need to be a fast learner." "If you're a good gamer, if you can jump in a game and figure it out really quickly, you're really good at this." "Coders get lost in the details." "Product people, people who are focused on solving a problem, on making money, they're going to be focused on marketing, they're going to be focused on user interface, they're going to be focused on all the right things." "I think this year it's gonna flip, and I think not having a coding background is gonna be more advantageous for the entrepreneur." @amasad with @jackhneel
English
@thsottiaux codex is the best subscription. got around $5k worth of free inference past couple weeks.
English
We don’t have evidence of a widespread issue with codex usage being drained faster than it should but there are enough reports and we have reset rate limits for plus & pro subscriptions while we work towards wrapping up our investigation over the coming 1-3 days.
Tibo@thsottiaux
We have found one issue that leads to some users seeing inconsistent usage across sessions but it is quite rare, affecting less than 1% of users. We are working on a mitigation and continuing the investigation. For the rest we are not seeing evidence of higher usage consumption other than the advertised token cost increase of GPT-5.4 being 30% higher than GPT 5.2 and GPT-5.3-Codex.
English
@ryancarson yeah, I did the same. I grew pretty happy about it, so I'm evaluating other options. Where did you go for similar service?
English
$200/mo ChatGPT Pro + gpt-5.4 xhigh + Codex Mac App is the most asymmetric upside I've ever seen in a tool chain.
A ridiculous cheat code for founders.
English
@MichaelArnaldi like with many hot takes, this is a consensus take.
there has to be something better than dinosaur UI right?
hot take is the opposite, thinking TUIs are the end game.
English
Hot take: TUIs are a phase, and GUIs that basically emulate TUIs are a phase too. We just don’t know how the IDE of the future will look like.
I am not complaining, I started programming in the terminal then it was notepad++ then richer and richer IDEs then finally VSCode
English
crazy how there’s still no clarity around this.
lots of great local first projects using agents sdk…nanoclaw, conductor, and now t3 code.
not being able to use subscription is a non starter for individuals.
Theo - t3.gg@theo
We want to add support for Claude via the Agent SDK so you can bring your subscriptions. We have a PR with the changes ready. We just don't know if we're allowed to ship it. The moment we get a 👍 from @trq212, @bcherny or @DarioAmodei, we will get this shipped.
English
@levelsio easy to dunk but icon had tier 1 execution except for the timing
English
currently whats the best AI code review tool? Greptile, bugbot, graphite, coderabbit? Something else?
English
/fast changing from 3x usage to 2x usage github.com/openai/codex/p…
English
fast mode is coming out of development in codex and will use 3x plan usage.
it directly compares to opus 4.6 fast.
unlike gpt-5.3-codex-spark, which was faster but weaker, with a smaller context window and no multimodal support.
English
If you have this level of moat, why in gods name would you give up 7% for $125k.
Paul Graham@paulg
In office hours with one startup today we came up with a plan that would continue to work even if the model companies had eaten all the other markets. That's a first.
English
GLM-5-Code just sitting on the pricing page
docs.z.ai/guides/overvie…
GLM-5 is the first open weight model i use daily. not quite good enough for my openclaw but maybe GLM-5-Code takes it over the edge.
English