benswerd

12 ms·

121.
by benswerd 10mo ago
thats the vibe i was going for
122.
by benswerd 10mo ago
Just pushed fix to typo — can you let me know which chromium browser on which platform? I'm primarily testing on Vivaldi on MacOS but also confirmed working on Chrome on MacOS.
123.
by benswerd 10mo ago
This is beautiful and I did not know it existed until now tx
124.
by benswerd 10mo ago
which browser are you using?
125.
by benswerd 10mo ago
What browser was it in?
126.
Drawing with zero-width characters(zw.swerdlow.dev)
122 points by benswerd 10mo ago|31 comments
127.
by benswerd 10mo ago
Figured this out a couple weeks ago, hoping it can lead to some cool new art.
128.
by benswerd 10mo ago
The tight click through for specific architectural features is sick
129.
Maxims of Good Software Design(kevingugelmann.com)
2 points by benswerd 10mo ago|0 comments
130.
Stringly Typed(hanselman.com)
3 points by benswerd 1y ago|0 comments
131.
by benswerd 1y ago
How do you think about remote configurability? Stuff like a lot of this needing to be A/B tested, models hot swapped, and versioned in a way thats accessible to non technical people? How do you think about this in relation to tools lik
132.
by benswerd 1y ago
> Maintain an LLM-friendly list of errors that LLMs commonly make when using your thing. For example, in Next.js the `headers` function, as of recently, returns a Promise (it used to return headers directly), and therefore you now have t
133.
<IsAgent/>(stytch.com)
4 points by benswerd 1y ago|1 comments
134.
by benswerd 1y ago
Stytch IsAgent is an interesting concept to me, moreso for the <IsHuman/> side of it than the <IsAgent/> side.
135.
by benswerd 1y ago
Which of these have you used and how are they useful to you? Do you think this is relevant at earlier stages of a project or only once you have tons and tons of docs? My instinct is that many LLMs.txt become less relevant over time as AI to
136.
by benswerd 1y ago
Our chatbot isn't trying to answer questions as much as it is to point people in the right direction. People have made chats that are better at actually solving problems but the usecase we targetted was I have a question how do I get t
137.
The missing trust model in AI tools(docs.freestyle.sh)
1 points by benswerd 1y ago|0 comments
138.
We revamped our docs for AI-driven development(docs.freestyle.sh)
90 points by benswerd 1y ago|35 comments
139.
by benswerd 2y ago
If I was less busy I wanted to try and make it run DOOM
140.
Reverse engineering OpenAI code execution to make it run C and JavaScript(twitter.com)
289 points by benswerd 2y ago|73 comments
141.
Show HN: AI-SQL, Tools for your AI to talk to your database(github.com)
4 points by benswerd 2y ago|0 comments