Benjie

That very closely aligns with my experience 👍

complex task I spend more time reviewing their code and checking it's right than I would have spent just writing it myself. That said, they're often great at then writing the test suite for the functionality :D

challenging their solutions typically get further and further from the ideal. If all you need is batch work that a junior could do then yeah they're a massive time saver, but if you're doing something innovative/complex, their results are frequently wrong in the most subtle of ways. I find for many

Right! What I mean by them being "terrible at code" is that they don't _understand_ what they're doing, they struggle with a-typical requirements. Like with a junior dev: they can string a bunch of docs/StackOverflow/etc together to produce something that seems to work, but the moment it gets

machine. Rather than having it mod my code directly, I prefer to have it generate a codemod that I can review, and then I run the codemod against my code.

Great for prototyping though, because prototyping is mostly autocomplete from READMEs/docs, and that’s where LLM’s shine. (v0 is great for UI!)

still useful in helping to shape the solution. Given the kind of code I’ve seen it produce, I’m loathe to hook it up to anything where `node --watch` or similar might actually execute the code and cause devastation, in a similar way that I wouldn’t allow a junior programmer to write code on my

Yeah, I use them heavily because they are useful for the drudgery. But you can never trust it, and once you get into complex areas (e.g. building fan-out with async iterables) it makes a hell of a lot of mistakes that are subtle to spot/track down. I throw away more code samples than I use, but it’s

ChatGPT gives me the best results in general, its tools are more reliable and claude and llama seem to go off the rails pretty quickly and immediately backpedal the moment you challenge them. I’ve not managed to run the 70b params llama, I don’t have enough VRAM or patience 🤣

In my experience it sucks at JS, but yeah I bet it nails the basics. Handling multi-consumer async iterables though and it's a) introducing race conditions, and b) telling me that my code has race conditions when it definitely doesn't 🙄

The worst thing is how confident it sounds.

I use LLMs heavily, and they're _terrible_ at code. Really really bad. Using them as autocomplete is okay-ish, but ask them to generate something serious or review something you wrote and they're wrong way more often than they're right.

I advise you treat them as the most junior of junior devs.

In exciting news, I will be on stage for the first time since 2018 at GraphQL Conf, but it wont be too nerve wracking as I'll have @benjie.dev with me. We've got a huge community update to present! graphql.org/conf/2025/sc...

Thanks so much to the GraphiQL maintainers and especially Dima for getting it over the finish line!

Ruru, our Grafast-enhanced GraphiQL distribution, is already updated and raring to go:

- Demo: grafast.com/myruru/#endp...
- Docs: grafast.org/ruru/

Made a quick video today with @jem.graphile.org to demonstrate a couple of the features in the new GraphiQL V5 with Monaco; it has familiar shortcuts, multiple cursors, command-palette to discover more capabilities, and even colour-matched braces to help you spot where your nesting went wrong!

EUROPEAN GRAPHQL-ers: help prove demand for events outside of San Francisco!

2025's GraphQLConf is in Amsterdam; attend, sponsor, share — everything increases the chance of a future return to 🇪🇺!

⏰ Early Bird ends in 4 days: get your ticket now! ⏰

The schedule is 🔥 graphql.org/conf/2025/sc...

I use Firefox primarily, but I often have to open Chrome to debug things because the Firefox experience isn't that great. Which is disappointing considering Firebug basically started the whole web devtools revolution! I loved that thing 💗

Thanks for trying Ricky 💪

graphql.org/conf/2025/

Just 10 days left to save $300 off your ticket to GraphQLConf '25 in **AMSTERDAM** 8-10th September. The talk line-up looks fantastic! Looking forward to seeing you there!

I don’t see many teams using GraphQL fragments, and I think they’re missing out.

Fragments let you co-locate your components data requirements with your components.

I wrote a quick post showing how this works to make maintaining your apps simpler.

Check it out: brookehatton.com/blog/enginee...

Really good article! One minor benefit that you might want to add: due to fragment masking it’s safe to remove a field your component isn’t using anymore - you don’t need to worry that some other component might be depending on it. This is critical for evolution without bloating!

Love that you’re writing about such an important topic! The link sadly is 404

London GraphQL July | Guild Jul 1st 6:00PM: Come and join us for the Summer edition of London GraphQL; Two talks on the most up-to-date GraphQL topics interspersed with

London GraphQL next week! New venue: American Express just across the road (literally!) from Victoria Station. Two great talks and a little recap of some of the things being worked on at the GraphQL working groups! Register so we know how much 🍕to get 😋

guild.host/events/londo...

I believe so; the plugin that integrates it has been updated I think!

Release graphiql@5.0.0 · graphql/graphiql We're thrilled to announce the release of the all-new GraphiQL 5, powered by the latest version of the Monaco editor! 🎉 We've also added support for comments in the Variables and Headers editors. T...

GraphiQL v5.0 was just released; it integrates Monaco editor (so you get coloured matching braces, multiple cursors, and all the stuff you're probably already familiar with in VSCode - press F1!) and also enables comments in the variables editor. Check it out! github.com/graphql/grap...

📣 The #GraphQLConf 2025 schedule is live!

🔥 Join us in Amsterdam 8-10 Sept. for deep dives and exclusive #GraphQL talks. Amazing sessions lined up!

➡️ graphql.org/conf/2025/

A candle in the shape of an asterisk with 5 burning ends

Just adding more ends to the metaphorical candle...

> Yours is fine as-is if you prefer the repetition for clarity.

ChatGPT with the sass! 😁

It really doesn't like that I never take its suggestions verbatim and prefer to make my own edits to my text.

I submitted two talks for this year. One for Graffle and one for a new project I recently started at @the-guild.dev. More to share about the latter in the coming weeks :))

Okay so yeah I missed the CFP deadline for GraphQLConf... but worry not - they extended it until Saturday and I submitted today!

You should submit too. There's only 48 more hours for you to do it, GET IT DONE! It won't be extended again. BTW: speakers get free attendance!

graphql.org/conf/2025/

It's really worth checking out the GraphQL JS documentation, a lot of effort has been put into renewing and expanding them.

Especially these new guides for GraphQL Errors and resolvers are absolutely top-notch!

www.graphql-js.org/docs/

Ashley Willis – ashley.dev Developer advocate, open source champion, and maker of things.

Spent 2 hours writing, 4 hours editing, and 2 days spiraling about whether anyone would care. finally hit publish anyway. ashley.dev/posts/fear-o...