🅰🅻🅸🅲🅴 (🌈🦄)
Unreasonably into rabbitholes. Software engineer.
Open to remote work in the EU!
| Blog | https://tsev.dev |
| https://www.linkedin.com/in/timseverien/ |
Tim Severien
@timsev
- 254 Followers
- 149 Following
- 1.4K Posts
Nik KantarI got a cold recruiting email from someone’s OpenClaw “agent”, and I don’t know if I’m supposed to appreciate that it clearly identified itself as such, but I’m definitely extra annoyed.
This next level “Dear {first_name}” waste of time can fuck right off.
@rik @LisaWesterveld Eens wat betreft de grootte, maar dan nog. Er zijn vele wegen naar Rome, en verschillende partijen willen verschillende routes bewandelen.
@crisverstraeten I've tried instructions in my prompt like "roast" and "grill", but in my experience, it tends to mostly change tone of voice, rather than truly scrutinising an idea. But I'll admit I haven't tried this with the latest flagship models due to budget constraints 😄
@LisaWesterveld Ik had het prettig gevonden als dit voorstel aangenomen was, maar een stem voor of tegen vertelt niet het hele verhaal, natuurlijk. Een stem tegen dit voorstel is niet per se een stem tegen beter naar kinderen willen luisteren.
Whilst LLMs are marketed to be perceived as intelligent, knowledgeable, helpful, and insightful, they're also commercial products that need to retain customers.
It's only logical that LLMs are extremely conflict averse and agreeable, even when we're wrong. Is that truly helpful and insightful? It reminds me of autocratic leaders where subordinates become agreeable pawns.
Maybe we shouldn't be autocratic leaders and put pesky humans in the loop.
@hdv This is nice and all, but where is the merch? I need this on a t-shirt to thwart NPC posing as colleagues!
Jokes aside, this is good stuff! I am curious: how come the benefits section structured into subsections, and the risk section is not?
@hdv what's the [___] setting in YouTube to get the [___] AI generated captions to contain [___] swear words that are in the [___] audio?
I've spent so much time on this article with WebGL demos, I'm fed up with it and just want to throw it in the bin 😅 How do people manage to write books?
@Anneke You happen to name things where we want to be consistent, deterministic, and (somewhat) transparent instead of being opaque and do wildly different things based on a randomly assigned seed or the tone of voice of your prompt 😅
The closest thing I've seen to what you're describing is GitHub's Agentic Workflows, which allows you to schedule an agent to complete a prompt. Could be interesting to update dependencies (and follow upgrade guides), for example.