Profile avatar
quentinfennessy.bsky.social
Looking for a vibrant AI community here instead of the other place. Also happy to engage with people in general. https://www.qfennessy.net/
42 posts 78 followers 521 following
Regular Contributor
Active Commenter
comment in response to post
Never heard of it
comment in response to post
It is really interesting that you could get that quality of a response from a 1B model. however I don’t think we’ve define what think means and how do we ask the question if we don’t know what think means?
comment in response to post
Looks good. Will you share your code?
comment in response to post
comment in response to post
comment in response to post
comment in response to post
comment in response to post
This sounds like excellent advice. I will try this soon. Thank you.
comment in response to post
Very very nice. Consider this reply as an homage.
comment in response to post
comment in response to post
I find it useful to consult with 4o, o1 and Claude
comment in response to post
OP: have I wandered away from your question?
comment in response to post
From Claude: "Software architecture gets better when you think beyond 3D: Feature space, scalability vectors, quality attributes, and testing dimensions create a richer solution space. Map your decisions across multiple dimensions to build more robust systems. #SoftwareArchitecture"​​​​​​​​​​​​​​​​
comment in response to post
Claude’s answer was too long for a single reply. I then asked: Summarize that response in a tweet​​​​​​​​​​​​​​​​
comment in response to post
I asked Claude: How can you use higher dimensions when creating and refining application architectures?​​​​​​​​​​​​​​​​
comment in response to post
This is a question I don’t understand. I will ask Claude.
comment in response to post
Also true for me although I’ve also been consulting with 4o
comment in response to post
*plonk* has been part of my approach to online communities since Usenet.
comment in response to post
Dogs are wonderfully unselfconscious
comment in response to post
I am. Imnsho if you are not you are doing it wrong.
comment in response to post
anaconda was actually a constrictor really enthusiastic about prepackaging numpy
comment in response to post
I'm a learner here. I think you are saying that BM25 can fill the function of vector dbs for RAG. is that correct?
comment in response to post
I would seriously like to understand your statement. No snark.
comment in response to post
This is a very cool article on LLMs and chess.
comment in response to post
Here's a game - find the LinkedIn article that wasn't created with a one paragraph ChatGPT prompt.
comment in response to post
That's so precious. 2026 for a conversational Siri. That Apple Car really slowed you down.
comment in response to post
I now expect it. Every day. When I need to solve a program, I no longer ask my imaginary friend Claude to write a python program. I ask for a python package with command line arguments, validation, and pytest and good test coverage.
comment in response to post
You suffer for your art
comment in response to post
Very nice! I asked about the arc benchmark. Answer:The Abstraction and Reasoning Corpus (ARC) is a benchmark designed to assess the general fluid intelligence of artificially intelligent systems, as well as humans (Moskvichev et al. 2023). …
comment in response to post
It is Little Bobby Tables all the way down.
comment in response to post
Surrender Dorothy!
comment in response to post
That’s exactly why I stayed on x. I hope the ai community bails from x completely. Thanks!
comment in response to post
OK now buying a Toyota is ON ThE TABLE!
comment in response to post
Sorry for your loss.