fsniper

Computers can code for you, but it's not a reason to forget about your mastery goals.

The printing has been around for ages now, still people master calligraphy.

Seeing many people using AI to prototyping and ordering boards, I am curios if JLCPCB, PCBWAY and the other prototyping services has upticks in their orders. It would be great to see some numbers.

I hope this is a joke and you really don't consider this as such.

It's in fact opening up too many opportunities to us humans. It's never been this easy to acquire information and accomplish things.

I have been using Gemini, Claude, ChatGPT, Grok (whatever is available) to create esp32 boards. As I am very new into this field, I mostly use them to query about schematics, ICs, routing, kicad, electronic foundations and so on. They are great help and provide me invaluable advice. However It's always better to validate one's result with others. I have many cases where they would provide me catastrophic advice and a second opinion fixed it.

It's always fun to get apologies from LLMS. I love it.

I'll check the newer models as soon as possible to see how much improvement they provide.

It's not a transplantation, as it implies a separate donor. Is it not?

Perhaps the title needs to clarify the source of the transplanted tissue as the same patient.

Well I have to agree on both counts.

Perhaps what "feel frustrating" needs to be agreed upon. I took it as I would not like to do this as I feel it to be kind of distasteful, not as I would take this but the code would be confusing.

That's my point. It should not be frustrating. It's part of the job. A responsibility a professional should be ready for.

On the part of vibe coded code, it can be a "bad code base" and readily so it's very understandable that reading it can be frustrating. However it could also be very well written. LLM written code can be considered a codebase that was written by a large portion of the developer community all together. It's not coming from some software. Software curated that from human written code one word at a time.

Examples:

  - When you are working in a team, you always take responsibility of the code your team mates has written, (review, maintain, bugfix, extend, refactor..)
  - Sometimes organizations reorg and you land in owning already written code by some other team (whole takeover)
  - Some other times you take a position to maintain and fix, refactor, bugfix an unknown codebase. You will work on this and leave. So it is never your code. (can be whole takeover, can be temporary ownership)

Perhaps they shouldn't be allowed to school until 16? School and other kids can expose theirs to many detrimental things.

"As a professional developer, I would find it very frustrating to have to debug someone else's code and also be held responsible for that code." Are you really a professional programmer? A very basic responsibility for many professionals are maintaining someone else's code.

For me it was related to some benign brain mass'es arm touching my ear nerve. This causes me some hearing loss, and also it was causing tinnitus. However I realized tinnitus directly correlated with my blood pressure. So If I control my blood pressure I don't have it.

AI writes faster, so the rate it incurs tech debt is proportionally higher. However it's ability to have large context kept on memory compared to humans is also a key component fighting against it. These are occasionally forgotten when code quality of AI for large code bases are discussed. So Yes I think humans also build spaghetti, but as they write slower they get to the same place a lot later. However humans can't correct it, or can't correct it fast enough. AI can.

Afaik there are already a few videos on YouTube.

That's not enough as the abstraction of hardware and software has increased exponentially. Not only that expectations are more. Just building for a definitive set of single configuration is not enough. Any software needs to run on Amd64, RiscV, Arm64, has to be containerised, and run on OsX, Windows, Linux.. Nowadays you would need more and more knowledge/experience to be able to function without external help.

I did vibe coded one, where It downloads images from my (currently defunct) immich server album, dithers them for 4 color inky display, downloads the weather data and display them. It's a fun little one. I wish the inky only had a "faster than 30secs full refresh" partial update capability.

Car manufacturer build infotainment never is.

Quality of YC investments are nose diving.

You can not trust human output without verification either. That's why you have tests, qa, staging envs, A/B tests..

I am with you on this. I am a good, and experienced infra engineer. And I feel like - no know that - llm can probably replace me if a good operator handles that.

The are lots of small "oh shit" moments for me. First interaction with an llm was already magical.

"This shit can emulate understand language, find a solution, answer it into words" .

Then came realisations it's not limited to single human languages, you can ask in one language and it could answer in another. It's also capable of understanding and generating code. Not only that, it's better than most humans for that. It can hear, it can see, it can paint, it can do music, it can sing.. It can combine, give a picture, ask for a music from that picture. Give a video, get software. It can mix and match.

After that came improvements, - no The revolutions - It started as a 4 year old with encyclopedic knowledge. It knew but could not convey, could not make sense sometimes. Was incorrect most of the time. Blubber. In a few years it matured to impeccable levels. It now can relate information with a lot of clarity, and it's less and less wrong. Nearly no hallucinations. It can do maths! Correct maths! Maths that I could not even my life depends on it. It's getting to a stage that it can proof where humans failed.

I am getting "oh shit moments" day by day.

You are paying restaurants for food to be prepared in the way you want. But this is not the same. Someone created some content the way they want. You haven't ordered that content. And you complain it's not prepared the way you like.

I have trouble understanding this. I don't see anyone complaining that we use microwaves and ovens instead of going for lit wood to cook or using search engines instead of crawling through libraries, or using Google Maps instead of using paper maps. These are tools. If output of an LLM conveys the ideas to be told, then what is the problem?

Not everyone needs to be magicians with language.

I have been a happy user of Gemini CLI.

What makes it "incomprehensibly bad." in your opinion?

What on earth? I can't understand USA at all..

I am very new to the crafts and I can attest trying to solder smd stuff, with correct equipment is way easier than soldering 2 wires correctly together. And I also kinda hate QFN like packages where you can’t see the pads.

I wish more IT companies were like Valve.

also kcp

Aren't we already at a worse place, where largest companies on earth doesn't have any support and you need to have a HN following to get their attention?

That depends on the kid and time. Mine carries the small cart and fills what we buy regularly. And other times we need to contrain him as expected.