HackerLangs
TopNewTrendsCommentsPastAskShowJobs

ixeption

no profile record

Submissions

[untitled]

1 points·by ixeption·bulan lalu·0 comments

[untitled]

1 points·by ixeption·2 bulan yang lalu·0 comments

Claude Opus 4.7 leaks system prompt randomly

old.reddit.com
2 points·by ixeption·2 bulan yang lalu·0 comments

[untitled]

1 points·by ixeption·2 bulan yang lalu·0 comments

Cognitive Debt: How AI-Generated Code Introduces a More Dangerous Kind of Debt

substack.com
2 points·by ixeption·3 bulan yang lalu·0 comments

Agentic Engineering: The good, the bad, the ugly

substack.com
2 points·by ixeption·4 bulan yang lalu·0 comments

[untitled]

1 points·by ixeption·4 bulan yang lalu·0 comments

[untitled]

1 points·by ixeption·4 bulan yang lalu·0 comments

comments

ixeption
·bulan lalu·discuss
I really think that this blog is part of the IPO story from Anthropic or OpenAI.

Because the reality is: A generalist can simply not guide the LLMs through the system nor can he/she verify the output. LLMs are force-multipliers and if there is nothing to multiply, it won't work.

As long we don't have AGI, humans will guide AI and if we get AGI, that's not something to even think of anymore. Software engineering was always about automation, if you want to be an artist, you better look for something else.