I was grinning ear to ear reading this, laughed together with a co-worker. What a brilliant, beautiful, thought provoking, ridiculous genius of a comedy.
Thank you, I felt both my intelligent and comic parts of the brain were hanging out in a bar.
Thank you so much for the kind words—this made my day. Would it be alright if I quoted this on the website? It really captures what the podcast aims for.
Our management introduced copilot last year, there was some mild hype, people were curious, gave it a spin, but it didn’t stick around in many conversations.
Now that everyone has access to Claude and claude-code, Copilot barely gets mentioned anymore. Maybe this wave dies down or they improve it, anyway these tools still have a long long way to go.
Finally someone said it, they're overconfident in their approach, don't consult us with the details of the implementation, they're trained to create mock APIs that don't follow structure, leading to lot of rework. The LLM actions should be measured, collaborative, ask for details when it's not present. It is impossible to give every single detail in the initial prompt, and a follow up prompt derails the train thought and design of the application.
I don't know if I'm using it right, I'd love to know more if that's the case. In a way the LLM should improve on being iterative, take feedback, maybe it's a hard problem to add/update the context. I don't know about that either, but love to learn more.
Most stacks now support some form of "plan" workflows. You'd want to first do this, and see if it improves your experience.
One workflow that works well for me, even with small local models, is to start a plan session with something like: "based on @file, and @docs and @examples, I'd like to _ in @path with the following requirements @module_requirements.md. Let's talk through this and make sure we have all the info before starting to code it."
Then go back and forth, make sure everything is mentioned, and when satisfied either put it into a .md file (so you can retry the coding flow later) or just say "ok do it", and go grab a cup of coffee or something.
You can also make this into a workflow with .rules files or .md files, have a snippets thing from your IDE drop this whenever you start a new task, and so on. The idea with all the advancements in LLMs is that they need lots of context if you want them to be anything other than what they were trained on. And you need to try different flows and see what works on your specific codebase. Something that works for projectA might not work for projectB ootb.
Also giving them more details seems to confuse them. There is probably a way around this, though. They are pretty good in finding a tiny silver of information out of the ocean. What I hate is that the industry is all geared toward the same model (chat bot). Imagine if we never invented the keyboard, mouse, GUI, touch screen, etc...
Yes, this is exactly why the "planning" approach never seems to work for me. Like every ounce of planning I do with the LLM it becomes a pound stupider at implementation time
Staples, Best Buy, CompUSA, Office Depot are not a monopoly, if you do not like one as a distributor, you can go sell at another, or you can very well sell at your own mini shop, as any commerce should be.
Best Buy/CompUSA/Office Depot employees do not scare the customers and block the customers coming to you directly.
These are the few problems of monopoly that Apple so conveniently takes advantage of.
The scare part refers to warning dialogs that iOS would pop up if you used a link to an external web browser to collect payments. It would warn you that Apple wasn't running the payments and use scary language warning you about potential fraud, etc. to try to scare people away. They would also demand a 27% fee from developers for collecting money on the web outside apps if users were paying from following that link.
As a developer, you are allowed to put only one 'link' (not a more user friendly button/supporting text) to your own website for any purchase, it has to be a small font, and not in the checkout flow that takes to Apple pay. Which in itself is very scummy to begin with.
To top it, when the user actually sees that hidden 'link' and clicks on that, the user is put up with a big screen of message that the external website doesn't have security, privacy etc.
If it is possible can you share an example/samples integrating the mentioned libraries. I’m starting to learn LiveView again, mostly want to use aggrid and other custom libraries. An article highlighting the possibility and pointers to samples would greatly help everyone in this space.
I don't have any examples from my code, so the best I can do is the relevant documentation.
Its all done using phx-hook https://hexdocs.pm/phoenix_live_view/js-interop.html#client-... . While the documentation makes it look a bit more fancy, it is just a JS object with mounted() function. I set up my columnDefs and gridOptions in that function, and add the handleEvent() callbacks that are where I get data from the elixir side using send_update/3. Ag-grids vanilla JS documentation has been very helpful for how to use it without the benefit of react/vue,
I was hoping I'm not alone, I had mixed signals reading the article. OP talks about how the interviews are harder in 2024 and says candidates are using AI to crack impossible time bound programming tasks and instead recommend companies to do the interview online. Somewhere something is disconnected
With all the turmoil in reddit, twitter and other social media, me and all of us here are so grateful there exists a forum which goes beyond the usual small talk, beyond known circles, with sincere and original thoughts no where else to be found on earth. A big thank you to all the mods and a very happy new year to everyone!
Thank you, I felt both my intelligent and comic parts of the brain were hanging out in a bar.