In our org (large US bank) there has been a huge rollout of github copilot and adoption has been very successful. For me it became essential to learn how this can help me in day to day coding/testing/devops etc. Right now I am churning out a python code to parse a csv and create a report. I have never learnt python before.
How to keep up with AI/ML as a full stack dev?
61–70 of 84 posts
Re: How to keep up with AI/ML as a full stack dev?
#62> I most often can’t see any use case for AI/ML I'm admittedly a skeptic on all this so take what I am about to say with a grain of salt: You should trust that voice. We're in a hype cycle. It was VR before and crypto before that. Big tech is trying _very_ hard to convince you that you need this. They need you to need this tech because they are lighting billions on fire right now trying to make it smart enough to do…
I do not understand the AI naysayers.
The other day I had an idea for a Chrome plugin. I'm a senior dev, but I've never made a Chrome plugin. I asked ChatGPT 4o if my idea was possible (it was) and then I asked it to create an MVP of the plugin. In 10 seconds I had a full skeleton of my plugin. I then had it iterate and incrementally add capability until it was fully developed.
I had to do some stylesheet tweaking and it asked for a permission that we didn't need, but otherwise it completely nailed it. Easily provided 95% of the work for my extension.
I was able to do in 60 minutes what would have probably taken several days of reading specs and deciphering APIs.
Is my Chrome plugin derivative? Yes. Is most of what we all do every single day derivative? Also yes.
How are people still skeptical of the value that LLMs are already delivering?
Re: How to keep up with AI/ML as a full stack dev?
#63Earlier quoted context omitted.
The question is what does programming with an LLM get you over batteries-included frameworks with scaffolding like Rails or Django? If the problem only requires a generic infra solution put together by an LLM instead of a bespoke setup, why not look into low-code/no-code PaaS solutions to start with? Unless the LLM is going to provide you with some uniquely better results than existing tools designed to solve the sam…
The question is what does programming with an LLM get you over batteries-included frameworks with scaffolding like Rails or Django? Three years ago an LLM would conversationally describe what the code would look like. Two years ago it might crib common examples with minor typos. Last year it could do something that isn't on StackOverflow at the level of an intern. Earlier this year it could do something that isn't on…
Absolutely. I posted a similar experience developing a Chrome extension with GPT 4o in a hour or so when it would have taken me at least a day to do on my own. I have no idea how people are hand waving LLMs away as no big deal.
I think the only justification for such a position is if you are a graybeard with full mastery of a stack and that's all you work in. I've dealt with these guys over the years and they are indeed wizards at Rails or Django or what have you. In those cases, I could see the argument that they are actually more efficient than an LLM when working on their specialty.
Which I guess is the difference. I'm a generalist and I'm often working in technologies that I have little experience in. To me LLMs are a invaluable for this. They're like pair programming with somebody that has memorized all of Stack Overflow.
Re: How to keep up with AI/ML as a full stack dev?
#64Earlier quoted context omitted.
My disappointment comes from understanding that what humans do is keystroke prediction. If the output that I want can be solved by the most likely next keystroke, then sure, that’s a good use case. I’m perfectly capable of imagining those cases. People who are all in on humanity seem to not get this and go wild. Don't mistake the "what" for the "how". What we ask LLMs to do is predict tokens. How they're any good at…
> Like previous GPT models, the GPT-4 base model was trained to predict the next word in a document… https://openai.com/index/gpt-4-research/ What humans do is materially different than that. When someone asks me a question, I don’t come up with an answer by thinking, “What’s the first word of my response going to be? The second word?…” I understand that the AI marketing wants us to believe there’s more magic than th…
> I don’t come up with an answer by thinking, “What’s the first word of my response going to be? The second word?…”
Actually, I'm not so sure that isn't exactly what we do. That's why it's called a "train of thought". You have a vague idea and you start talking and lo and behold out comes a pretty coherent encapsulation of your idea that is informed and bounded by the token relationships of your language.
Try answering a question with the order of your sentence reversed and you'll find it damn difficult. That answer of yours is not completely well formed just waiting for your mouth to get it all out. You're coming up with the answer one token at a time.
Re: How to keep up with AI/ML as a full stack dev?
#65> I most often can’t see any use case for AI/ML I'm admittedly a skeptic on all this so take what I am about to say with a grain of salt: You should trust that voice. We're in a hype cycle. It was VR before and crypto before that. Big tech is trying _very_ hard to convince you that you need this. They need you to need this tech because they are lighting billions on fire right now trying to make it smart enough to do…
I do not understand the AI naysayers. The other day I had an idea for a Chrome plugin. I'm a senior dev, but I've never made a Chrome plugin. I asked ChatGPT 4o if my idea was possible (it was) and then I asked it to create an MVP of the plugin. In 10 seconds I had a full skeleton of my plugin. I then had it iterate and incrementally add capability until it was fully developed. I had to do some stylesheet tweaking an…
It's probably because it's providing different amounts of value to different people. For some people, it's not giving any benefits, and in fact making their work harder (me). They are skeptical because people naturally don't believe each other when their personal experience does not match up with another.
Re: How to keep up with AI/ML as a full stack dev?
#66Earlier quoted context omitted.
> Like previous GPT models, the GPT-4 base model was trained to predict the next word in a document… https://openai.com/index/gpt-4-research/ What humans do is materially different than that. When someone asks me a question, I don’t come up with an answer by thinking, “What’s the first word of my response going to be? The second word?…” I understand that the AI marketing wants us to believe there’s more magic than th…
> I don’t come up with an answer by thinking, “What’s the first word of my response going to be? The second word?…” Actually, I'm not so sure that isn't exactly what we do. That's why it's called a "train of thought". You have a vague idea and you start talking and lo and behold out comes a pretty coherent encapsulation of your idea that is informed and bounded by the token relationships of your language. Try answeri…
Like Heptapod B. The "next word" argument is pervasive and disappointingly ridiculous. If you present an LLM with a logic puzzle and it gives you the correct answer, how did it "predict the next word"? Yes, the output came in the form of additional tokens. But if those tokens required logical thought, it's a mistake to see the what as the how.
Re: How to keep up with AI/ML as a full stack dev?
#67> I most often can’t see any use case for AI/ML I'm admittedly a skeptic on all this so take what I am about to say with a grain of salt: You should trust that voice. We're in a hype cycle. It was VR before and crypto before that. Big tech is trying _very_ hard to convince you that you need this. They need you to need this tech because they are lighting billions on fire right now trying to make it smart enough to do…
I do not understand the AI naysayers. The other day I had an idea for a Chrome plugin. I'm a senior dev, but I've never made a Chrome plugin. I asked ChatGPT 4o if my idea was possible (it was) and then I asked it to create an MVP of the plugin. In 10 seconds I had a full skeleton of my plugin. I then had it iterate and incrementally add capability until it was fully developed. I had to do some stylesheet tweaking an…
It's the best API searcher ever made but most people don't search APIs. They are waiting for it to make them a grilled cheese or something.
Re: How to keep up with AI/ML as a full stack dev?
#68There's two sides to this: using AI/ML/LLMs to augment your own development ability, and using AI/ML/LLMs within apps you build.
For the former side, Copilot-type implementations are pretty intuitively useful when used this way. I find it most useful as an autocomplete, but the chatbot functionality can also be a nice, slightly-better alternative to "talk to a duck when you're stuck." That said, I'll focus on the latter side (using AI/ML in your actual work) from here.
Generalized AI/ML/LLMs are really just a "black box API" like any of the others in our toolbelt already, be they Postgres, Redis, SSE, CUDA, Rails, hell, even things like the filesystem and C atop assembly. We don't need to know all the inner workings of these things, just enough to see how to use the abstraction. You probably take when to use a lot of these things for granted at this point, but the reason we use any of these things is that they're good for the specific problem at hand. And LLMs are no different!
What's important to recognize is the types of problems that LLMs are good for, and where to integrate them into your apps. And, well, a pretty obvious class of this is parsing plain text into structured data to be used in your app. This is pretty easy to prompt an LLM to do. OpenAI and WebLLM provide a pretty straightforward common set of APIs in their NPM libraries (and other language bindings are pretty similar). It's far from a "standard," but it's definitely worthwhile to familiarize yourself with how both of these work.
For an example, I've made use of both OpenAI and WebLLM in an "Event AI" offshoot to my social media app [1], parsing social media events from plaintext (like email list content, etc.); feel free to test it and view the (AGPL) source for reference as to how I'm using both those APIs to do this.
For projects where you actually have money to spend on your LLM boxes, you'll probably do this work on the BE rather than the FE as demoed there, but the concepts should transfer pretty straightforwardly.
If you're interested in really understanding the inner workings of LLMs, I don't want to discourage you from that! But it does seem like really getting into that will ultimately mean a career change from full-stack software engineering into data science, just because both have such a broad base of underlying skills we need to have. I'm happy to be wrong about this, though!
[1] Source: https://github.com/JonLatane/jonline/blob/main/frontends/tam... | Implementation: https://jonline.io/event_ai
Re: How to keep up with AI/ML as a full stack dev?
#69Earlier quoted context omitted.
I do not understand the AI naysayers. The other day I had an idea for a Chrome plugin. I'm a senior dev, but I've never made a Chrome plugin. I asked ChatGPT 4o if my idea was possible (it was) and then I asked it to create an MVP of the plugin. In 10 seconds I had a full skeleton of my plugin. I then had it iterate and incrementally add capability until it was fully developed. I had to do some stylesheet tweaking an…
It's the best API searcher ever made but most people don't search APIs. They are waiting for it to make them a grilled cheese or something.
It's the best API searcher for APIs which are used a lot. If you want do anything other than the most common thing it can be worse than useless. (I've been running into this in a project where I'm using Svelte 5, and the language models are only interested in telling me about/writing Svelte 4, and transformers.js, where they tend to veer off towards tensorflow.js instead. This despite me explicitly mentioning what version/library I'm using and the existing code being written for said version.)
Anyways, they can definitely be very useful, but they also have a golden path/winning team/wheel rut effect as well which is not always desirable.
Re: How to keep up with AI/ML as a full stack dev?
#70Earlier quoted context omitted.
I have not seen evidence of LLM use making programming way faster. Both in my own work, or from the work of others who make this claim.
Then you don't understand how to use the tools. LLMs are an accelerator for people who learn how to work with the prompts correctly and already have a good grasp of the domain in which they are asking questions.
Can you point me to a tutorial or tutorials that clearly show the claimed effectiveness?