That sounds crazy to me, Claude Code has so many limitations.
Last week I asked Claude Code to set up a Next.js project with internationalization. It tried to install a third party library instead of using the internationalization method recommended for the latest version of Next.js (using Next's middleware) and could not produce of functional version of the boilerplate site.
There are some specific cases where agentic AI does help me but I can't picture an agent running unchecked effectively in its current state.
Indeed. Attaching the link (of the correct page) of the documentation worked in this case but I would've been faster than the AI. LLM.txt has been hit or miss. Maybe I need to adapt my workflow and have a granular plan of what needs to be done.
However the complexity is in knowing what to do and when. Actually typing the code/running commands doesn't take that much time and energy. I feel like any time gained by overusing an LLM will be offset by having to debug its code when it messes things up.
I have seen it doing incredible stuff. One shotted adding a feature that included modifications to a proprietary backoffice system, db schema updates, defining new api models, implementing changes on the backend and then on the frontend.
I've also seen seen it choking when tasked to add a simple result count on a search.
Is it cheap? It adds up really quickly. One shot at trying to build an iteration of a simple python app (<1000 LOC tops) can cost between $1 and $5. And that’s a single attempt.
And this is just the tip of the tip of the iceberg of what even a medium sized startup spends. This is not cheap in any way.
This is where prompting comes in. You need to remember to tell it about which libs you want or encourage it to web search to find the latest ones, or use something like context7 MCP to get the latest versions.
Claude is always a little behind latest versions because of knowledge cutoff. Also I know the i18n lib you're talking about and it was probably the right call.
You could've stopped your sentence at "I don't know how a country filled with guns can survive."
The main downside of abusing the words nazi and fascist is that it gives an out to the actual fascists out there. When it comes to gun violence, there are a lot more (self proclaimed) neo-nazis killing innocent people than people killing them.
Fully agreed. I am however worried by the fact that Firefox is basically kept alive by Google. I assume it's just so that they can pretend Chrome isn't a monopoly, but the minute Firefox becomes an inconvenience they can stop financing it. I hope we can find a way for Firefox to sustain itself long term.
It’s a valid concern, and it may not be possible to properly address so long as Mozilla in its current form continues to be the controlling party of Firefox/Gecko. The best scenario might actually be for Mozilla to collapse and some other NPO or PBC with better financial sense to pick up the projects and their engineers.
Google pays Firefox for traffic acquisition, not out of pity. If Google stopped paying, another search engine like Bing or Perplexity would be happy to take over.
True, but what happens when Firefox's marketshare decreases to the point where the amount of traffic lost by not having the Google deal stops mattering to Google?
If Google does the math one day, and determines that they won't lose out anymore by not paying Firefox they'll stop paying.
It's revenue share based, so the cost to google is the time it takes to renew the deal. This is a fixed cost that doesn't depend on the market share of Firefox.
I've also made the move to Zen. I think Arc users will feel right at home there. It hasn't quite reached he same level of polish just yet but being in active development is a big plus.
On top of that, Zen can be personalized with CSS. As someone who spends a lot of time in the browser, it's been awesome to be able to tailor it to my needs.
https://docs.zen-browser.app/guides/live-editing
Slight tangent but while this is bad with disclosed ads, I think it's even worse with algorithmic feeds. A lot of users don't realize that not only their feed is heavily personalized, but the "top comments" are also selected just for them.
This bias will be implemented in LLMs sooner or later. Combined with the current misunderstanding of "AI" by the general population, it makes me worried about the future of misinformation.
Cambridge Analytica/SCL/Meta are knowingly in the personalised propaganda business - political behaviour modification tailored to individual emotional triggers, sweetened with personally irresistible content that looks harmless but exists to hook users on the real payload.
"And here's some more content about your favourite band/movie/TV show/influencer. And now, some reels. Aren't those cats cute? But anyway. Here's a reel that looks unobjectionable but subtly makes a political point you should definitely think about. Would you like to know more?"
AI has the potential to make this much more effective.
Certain noteworthy historical figures from Germany would have been absolutely delighted.
It took me a while to make the switch from copy pasting in a chat to Cursor, but I do see an improvement in my productivity in a couple of ways. For context, this is doing front-end development and knowing what I'm doing. The benefits aren't there if I'm using a technology/library I'm not familiar with.
For the most part, I just type code the same way I use to but I get:
- an auto-complete on steroids
- the tab feature reminding me of impacted code I forgot to update after making a change elsewhere (big one as I easily get distracted).
I very rarely use the chat/composer. Usually I'm faster by going through files manually and making changes myself helped by the features mentioned above.
This is one of those times where the shared article isn't very insightful but the topic and discussion that ensues are interesting.
I've had Prime every now and then to watch some shows. I've now stopped using it entirely because I don't want to support the company. Unsubscribing within the Prime Video app was a horrendous experience where I got stuck in a loop. I've also been re-subscribed without my knowledge and the only reason I can find to explain this is that opening the app on my TV and clicking "ok" once re-subscribes me. I have since removed my payment methods from the site.
As an e-commerce platform, it's been going downhill for a long time. Now it just feel like a more expensive Temu/Wish.com. Full of dark patterns, low quality items, and cheating 3rd party resellers.
It's got more dark patterns than Temu. So many sneaky patterns to try and trick you in to signing up for Prime. I've never had an issue with Temu - not somewhere to get name brand products, but if I want something cheap and don't mind waiting a couple of weeks then it works better than Amazon.
Amazon basics is so hit or miss anyway. I bought a bunch of their rechargable batteries, but it's hard to discern the quality of invisible specifications.
I kind of wonder if they suck, because some physical amazonbasics things I've bought were actually not great.
Counterfeit meaning not the product they actually paid for.
For example, an RTX 3050 probably had thing margins when it came out (or some equivalent GPU with low margins).
But! You can still counterfeit it and put in a GT 430 and have the same shroud - and people do just that.
Amazon HDMI cables are cheap, I’m sure, but an older spec HDMI cable with no HDCP is probably even cheaper. And you can’t really tell the difference… until everything is streaming at 480p.
I'm sure some might be acting in good faith. Nonetheless, I believe they're generally wrong. From experience going to the office for the sake of going to the office doesn't help.
The company I worked for who had the best company culture was fully remote but put a strong emphasis on communication.
Meanwhile, my current company insists that people show up to the office regularly, and it's costing me 8h per week and I get a less comfortable work environment. It does nothing to solve our communication issues however. Even worse, it feels like some execs think the company culture will build itself just by putting people together in a room. It reminds me of people who schedule meetings because they don't know how to organize their thoughts and write down what they want to say.
edit: Rephrasing, I got emotional. I don't know how you managed to have a life with three kids, night school and a full time job + commute. All my free time goes to my child and family. I barely have time for hobbies. If a company wants to take more of my time they better have a compelling reason.
Last week I asked Claude Code to set up a Next.js project with internationalization. It tried to install a third party library instead of using the internationalization method recommended for the latest version of Next.js (using Next's middleware) and could not produce of functional version of the boilerplate site.
There are some specific cases where agentic AI does help me but I can't picture an agent running unchecked effectively in its current state.