I replaced a $120/year micro-SaaS in 20 minutes with LLM-generated code(blog.pragmaticengineer.com)
blog.pragmaticengineer.com
I replaced a $120/year micro-SaaS in 20 minutes with LLM-generated code
https://blog.pragmaticengineer.com/i-replaced-a-120-year-micro-saas-in-20-minutes-with-llm-generated-code/
4 comments
Have a similar experience with replacing Copilot after looking at the system prompts and wtf'n
It's certainly taken longer because it's more involved, certainly less than a month for someone already familiar with vs code extension development. A weekend project for the basics level like OP.
I hope Ai shakes up the economy in a good way, away from consolidation, because it's not working for most of us right now.
It's certainly taken longer because it's more involved, certainly less than a month for someone already familiar with vs code extension development. A weekend project for the basics level like OP.
I hope Ai shakes up the economy in a good way, away from consolidation, because it's not working for most of us right now.
Does Copilot include system prompts at the extension level or the API level?
Copilot prompts are in the extension, they make the system_prompt field you send in the API request to an LLM
The underlying models have various guardrails and alignments that you cannot work around trivially
For Copilot, you can look at the code on GitHub, it is JSX based, which is interesting, they pass context budget info around
For Claude Code, and many others, you can find the extracted prompts online
At this point, they are all dynamically generated from fragments and contextual data (like what files or language you're working with)
The underlying models have various guardrails and alignments that you cannot work around trivially
For Copilot, you can look at the code on GitHub, it is JSX based, which is interesting, they pass context budget info around
For Claude Code, and many others, you can find the extracted prompts online
At this point, they are all dynamically generated from fragments and contextual data (like what files or language you're working with)
I think the article is right that a lot of people will replace such things with LLM-generated options, but I also think a lot of people won’t. For programmers and technical types, maybe so. But I suspect that for a lot of nontechnical users, even the relative ease of AI coding will remain more than they really want to do.