I see LLM usage as more of a spectrum than a binary decision. For the most critical parts of my day job, I still don't commit anything from an LLM that I haven't checked myself. My brain has never felt busier and more engaged on that stuff. But then I also produce absolute mountains of code (side projects, internal tooling) that I don't have time to review and frankly it doesn't really matter what the code looks like. You can choose how much to engage your brain. Every project is different.
I might try this but I've gotten so accustomed to talking to agents as I would a human - I worry that if I get accustomed to speaking coldly and directly to agents I'll find myself talking like that to people.
I think that's something that has changed in the last 6 months. It used to be standard practice to let Claude scan everything once and describe your repo in the CLAUDE.md - the logic being you don't want to do that every time for every prompt. But now I think the tooling has gotten so good at just grepping around your repos, and maintaining memory from previous sessions, that the best practice is only use CLAUDE.md for things that aren't obvious from the code.
Right now I feel like I'm getting so much value out of working daily with Fable, but I'd be embarrassed to share my sessions because they're pretty much how I would talk to a colleague. Spelling mistakes and half-baked thoughts included. Gone are the formal sounding prompts and specs I was proudly writing 6 months ago. But it's getting the results.
To be clear I don't want to detract from what you've made here. Agentic engineering done right does still require a lot time and attention IMO. But you've still got a challenge on your hands to demonstrate to potential adopters that you're in it for the long haul.
I think a challenge for a vibe coded library to gain adoption is if there is a lack of human time investment in its creation, how do we know there will be investment in its maintenance?
I know plenty of women who have delayed starting a family and they were well aware of the increased risks.. as they struggled to meet the right person, establish themselves professionally etc. They were certainly not brainwashed.
We even find ourselves creating PRs in situations where the code is going to be merged immediately anyway, and tagging other devs, just so they have a convenient way to see what got merged and why. So people don't lose track of what is in the codebase.
I'm back in the UK for a bit and one of the first things I notice besides the infuriating and pointless "Accept Cookies" popups on every webpage is that I can't view half the news about Palestine on Twitter/X without handing over a government ID to prove I'm an adult. No thanks. Please let's find a way to protect children without giving up the right to anonymity on the internet.
Now feels like a very good time to be a small team of experienced developers who can largely work on stuff by themselves and not a corporation of hundreds of developers of varying abilities all now trying to show how much code they can generate and how many tokens they can burn.
I think one problem is that a lot of sleep doctors are essentially CPAP salespeople and they will just keep pushing that even if you protest that you don't feel any better. I got better answers from an ENT doc who did a Drug-Induced Sleep Endoscopy and told me mechanically why I was not breathing well at night.
IMO if you've got a use case that requires querying in so many ways that you need several indexes, then DynamoDB is probably the wrong choice. It excels at stuff like user specific histories that are well partitioned, read back in one way, and ideally can be written asynchronously by a separate writer process.
So excited to play this but also sad that they didn't get the IP to use Kerbals. Not loving the cutesy baby kittens.. should have gone with a more serious animal like capybaras.
Yup.. just started seeing OAuth errors in the desktop app. This is becoming an almost daily reminder of the risks of being locked into one AI platform.
I totally get this and I also think it's now the case that making a PR of any significant complexity, for a project you're not a maintainer of, isn't necessarily giving that project anything of value. That project's maintainers can run the same prompts you are running - and if they do, they'll do it with better oversight and understanding. If you want to help then maybe's it's more useful to just hashout the plan that'll be given to an AI agent by a maintainer.
I don't buy this journey vs destination binary I keep hearing. I always considered myself lucky to be 44 and still writing code all day. I love the journey - the mental satisfaction of creating something complex yet elegant. The perfectionism that leads you to ask yourself can this be simpler, faster etc. But I also now love creating things that frankly I was never going to find the time for.
As much as I'm enjoying all these articles about bad MacOS UI design.. people really be refusing to upgrade over this? I'm sitting here on Tahoe happily resizing windows all day (the cursor change lets you know you're grabbing the correct part).
It's really not and I don't think it's worth arguing with you but.. Zionism is the establishment of a Jewish majority state.. which requires the expulsion of much of the existing non-Jewish population. That not the same as other countries acknowledging the borders of Palestine.
Trying to frame the violent expulsion of Palestinians from their homeland as just "Jews trying to live in their own homeland".. isn't working in 2026 and nobody needs to read the thoughts of a man who saw Cecil Rhodes as a kindred spirit.
At analyzing and reproducing language.. words, code etc sure because at their core they are still statistical models of language. But there seems to be growing consensus that intelligence requires modeling more than words.