-
Notifications
You must be signed in to change notification settings - Fork 7
Add "building coding agents" chapter #13
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
hey @automata don't we already have the mention to https://ampcode.com/how-to-build-an-agent for this in the introduction? |
Hi @glumia, yes, that's a great blog post! However IMHO there's a lot more to talk about, specially considering different LLMs, multiagents, etc. So the goal here is to start writing more deeper "chapters" investigating topics like that. I believe we can also have "chapters" to pave the ground for people that came into contact with AI coding (and programming in general) through vibe coding. For instance, someone who just "vibe coded" its way and created a new webapp, it would be interesting to this person to know more about web development, frameworks, how to debug, etc. What do you think? |
gotcha, I'm only afraid that we lose the original spirit of the repo (a roadmap to start coding with AI) by trying to cover too many things (i.e. DOTADIW); imo the web is already filled with plenty of blog posts and articles that cover those topics, I would personally see little value if they were rehashed here (nowadays it's easy to research anything and find relevant sources of information using the LLMs themselves) PS: that's just the opinion of a random person on the internet, feel free to proceed as you prefer of course (don't really wanna come and annoy you on the PRs) |
@glumia ah I truly consider your opinion, I really do! :-) I believe you're right, that's why I'm keeping this as a draft PR for now, let's see how it will evolve. Maybe it will make sense just like it happened with the llm-course one. But I get your point! I really want to keep it as a roadmap. Given that, may I ask you what you think we should cover more on the roadmap itself? Maybe not the best place to discuss it, but I feel that you have great ideas to share! Tks! |
@automata yea, would be great to have a chat of some sort (discord/slack/whatever), maybe we could use github discussions but I never used these tbh. I would keep it focused on what's the best way to use LLMs for coding, and update it from time to time as better tools and models get released or we discover new methodologies to more effectively work with them. E.g. yesterday I was noticing how aider is being left behind by cursor and claude code because they've access to a wider set of tools (most notably searching for text in the repo), this could be something to mention looking for when picking an AI coding assistant. In general I'm for less rather than more, ideally the roadmap would be short enough to allow anyone to quickly get up to speed, and then contain a list of carefully selected resources to go deeper into any specific topics (that's the thing I struggle with nowadays, it's not about finding information but the time it takes to discern signal from noise). |
Add a chapter (or maybe an independent sub-guide?) demystifying how AI coding agents work under the hood.