Bottom-up programming as the root of LLM dev skepticism(klio.org)
klio.org
Bottom-up programming as the root of LLM dev skepticism
https://www.klio.org/theory-of-llm-dev-skepticism/
8 comments
This is a novel point for me and does seem to make sense. I'm definitely more of a "bottom up" programmer and haven't really been able to vibe (pun intended) with LLM workflows so far. In the cases where I can do a more top down approach (usually small, self contained projects) they work much better.
Hrrrm, nope, can't be bottom up programmers either. I'm a bit more towards the bottom-up exploratory style at the moment, and have quite a lot of fun with Opus 4.5 providing power steering.
I should probably try a test project top-down to see if I can get more out of it though.
Either way, if you just sit on your hands and expect the LLM to magically do all the work for you, you're a little bit mistaken, yet. (With certain exceptions proving the rule) .
I should probably try a test project top-down to see if I can get more out of it though.
Either way, if you just sit on your hands and expect the LLM to magically do all the work for you, you're a little bit mistaken, yet. (With certain exceptions proving the rule) .
My experience is exactly the opposite. With AI, it's better to start small and simplify as much as possible. Once you have working code, refactor and abstract it as you deem fit, documenting along the way. Not the other way around. In a world abound of imitations and perfect illusions, code is the crucial reality to which you need to anchor yourself, not documents.
But that’s just me, and I'm not trying to convince anyone.
But that’s just me, and I'm not trying to convince anyone.
in my experience you do both. small ai spike demos to prove a specific feature or logic, then top-down assemble them into a superstructure. The difference is that I do the spikes on pure vibe, while reserving my design planning for the big system.
Show me the working code and the working product. Anything less is just another blob of PR nonsense, human or LLM generated.
You can do that yourself with opencode and a $10-15 subscription to a good model.
I used to be a skeptic, but this month the pairing of tooling and models I discovered is finally good enough to build a static Go app with a TypeScript front-end, along with all the deployment and compile steps, despite knowing nothing about either of these.
Insufficient or bad wrapper will hobble a model. An insufficient model will mean the wrapper cannot help much. Both have to be in place, and this month they finally came together for me to create ready-to-go stuff.
I used to be a skeptic, but this month the pairing of tooling and models I discovered is finally good enough to build a static Go app with a TypeScript front-end, along with all the deployment and compile steps, despite knowing nothing about either of these.
Insufficient or bad wrapper will hobble a model. An insufficient model will mean the wrapper cannot help much. Both have to be in place, and this month they finally came together for me to create ready-to-go stuff.
Another classic AI fan article: thin veil of reasoning to finally go back to "because they are stupid".
With all due respect, did you read the article... ? I'm not sure how you take that away from it at all. It says rather explicitly that both top down and bottom up are valid approaches.