scastiel

Interesting, I did and document the same kind of experiment a few months ago [1], it looks like so much changed since then!

[1] https://betweentheprompts.com/40000-feet/

Totally agree, and I go even farther: I consider that the PR author is totally responsible of merging their PR or not, and so no comment is blocking. I can make a comment I consider blocking, but there might be very good reasons that I can't see to still merge the PR. Fixes can come in a follow-up PR. As long as everyone takes responsibility for their work, it is a well-working system.

This will get upvotes just because the link is too small and close to the upvote button…

Indeed, switching partially from Cursor to Claude Code increased the bill by a lot! Fortunately I use Claude Code mostly at work and I had no trouble to convince my boss to pay for it. But I’m still not sure how I’ll continue building side projects with Claude Code. Not sure I want to spend $20 each time I want to bootstrap an app in an evening just for fun…

I agree, prompt engineering really is the foundation of working with AI (whether it’s for coding or anything else).

For the first time in my career, I have no idea what software engineering will look like in 5-10 years. This post documents the 2-hour experience with Cursor that shifted me from AI coding skeptic to daily user.

Running a database accessed that many times on a $4 Digital Ocean droplet? I'd be very curious to see that ;)

The web hosting costs basically nothing. Most of the cost comes from the database.

For my expense sharing app [1], I added receipt scanning [2] in a few minutes and a few lines of code by using GPT 4 with Vision. I am aware that LLMs often are a solution looking for a problem, but there are some situations where a bit of magic is just great :)

It is a Next.js application, calling OpenAI’s API using a plain API route.

[1] https://spliit.app

[2] https://spliit.app/blog/announcing-receipt-scanning-using-ai

This feature was suggested here: https://github.com/spliit-app/spliit/issues/14

It doesn’t look like an urgent feature to implement, but it’s on the plan ;)

Yes, or on Twitter, LinkedIn…

Indeed that would be nice, but the main use case I had in mind is about sharing on social media, and as far as I know they don’t support animated SVG…

Glad you like it '^^

Yes, I just changed it, makes more sense :)

Yes I’ll add more information about the project: example output, how it works, tech stack, etc.

Haha I am not sure it would be that nice to have a video that lasts 2 minutes to display GitHub stars ;)

So useless, I love it XD

You don’t become a good software engineer because you have strong opinions.

You become a good software engineer despite having strong opinions.

The web development might not be perfect, and yes it’s evolving very fast, but it is not worse than it used to be.

I’m a bit disappointed in the article. With the title I expected to read something about development that is not web (system, embedded…), and it made me curious. Sadly it isn’t; the author is criticizing an ecosystem they left a while ago, and regret not to be able to keep up with it. Pretty normal, isn’t it?

It’s very easy not to be able to keep up with something you’re not a part of anymore. We shouldn’t expect things to stay the same forever. It’s okay to take a step back and reflect on how things are (yes, maybe web dev becomes too complicated in certain ways), but it’s hard to judge when you’re completely out of it (the author stopped web development “in 2009”).

Is it really Stripe? If so, WHY not using the Stripe domain name and make us triple check that the website is legit? :(

It seems to be an example website, but with my payment information… Could someone have made this website just to get payments from people thinking it’s a legit example page from Stripe?

Wow thanks for this amazing and useful feedback! I’m glad you liked the book, and will take your suggestions into account :)

When I became an engineering manager I read the book The Manager's Path by Camille Fournier. I found it very helpful.

Hi! I tried to list all the advice I gave and received about Pull Requests and Code Review into a short (37 pages) and free guide, with real-life examples and actionable insights.

Just had a closer look, it seems the second question’s answer is: creators use their own Stripe account, so the problem should not happen here.

Nice platform it seems, but I have a couple of questions:

- If it is free, does that make me the product? How will you make money?

- How do you prevent the problems that Flurly had, which led to Stripe closing their account?

I’m tweaking a workshop I give to beginners who want to learn React & Next.js. I like thinking about the things I can improve to make the learning experience as good as possivle!

+1

I would even add a “I’m feeling lucky” button, to redirect to a random blog ;)

https://scastiel.dev

I mostly post about frontend development: React, Next.js, TypeScript…

I guess one of the reasons they’re hard to convince is that they don’t understand why they should do it. Hopefully my post contains some tips to make them realize how it can help them and the team.

But ultimately, it can be a team rule to require every developer to spend some time on code review each day. I worked in some teams where you needed to review PRs to get review on your own. It worked pretty well :)

On this point I agree with you. And in my experience, it happens as often with seniors as with juniors. The famous “LGTM” on a 1000-line PR ;)

Thank you, your point is the exact one I argue against in my post, which makes me happy I wrote it. No, junior developer reviews AND approvals are not a waste of time. If your code can’t be understood by a junior, it shouldn’t be considered as good enough to be merged.