swyx

15 ms·

331.
by swyx 9mo ago
> a `.network()` method that turns any agent into a routing agent say more pls?
332.
by swyx 9mo ago
> That last one took a bit of time, we went down the ESM/CJS bundling rabbithole, ran into lots of monorepo issues, and ultimately opted for a more explicit approach. shudders in vietnam war flashbacks congrats on launch guys!!! f
333.
by swyx 9mo ago
ooo i love this term. does it have an origin? related meme i saw today: “bro I spent all weekend in Claude Code it’s incredible” “oh nice, what did you build?” “dude my setup is crazy. i’ve got all the vercel skills, plus custom hooks for e
334.
by swyx 9mo ago
ooh, LLM Recsys alert! (we had an LLM Recsys track at ai.engineer last year). official announcement here: https://x.com/XEng/status/2013471689087086804 looks like this is the "for you" feed, once again s
335.
by swyx 9mo ago
Why NanoLang? NanoLang solves three problems: LLM Code Generation - Unambiguous syntax reduces AI errors Testing Discipline - Mandatory tests improve code quality Simple & Fast - Minimal syntax, native performance Design Philosophy: Min
336.
by swyx 9mo ago
> HN is obviously very pro-ai mm i'd actually say its more moderated, lots of cynics/skeptics.
337.
by swyx 9mo ago
maybe throw it in your friendly neighborhood LLM
338.
by swyx 9mo ago
> single keystroke could move the cursor halfway across the file to exactly the right spot. sorry what is this? exaggeration?
339.
by swyx 9mo ago
semi relatedly i stumbled upon Dead Planet Theory a while back and it stasy rent free in my head. https://arealsociety.substack.com/p/the-dead-planet-theory
340.
by swyx 9mo ago
> So, much like with a public git repo, you should be comfortable with the fact that anything you put there is public and will get indexed. whats the sota on atproto encryption dan? just publish encrypted stuff with sha 256 and thats it?
341.
by swyx 9mo ago
> Apps may come and go, but files stay—at least, as long as our apps think in files. yes: https://www.swyx.io/data-outlasts-code-but all lasting work is done in files/data (can be parsed permissionlessly, still usef
342.
by swyx 9mo ago
hi! :) thanks for supporting my work. ok yea that makes sense - i was a langfuse only user but sounds like i might want to check out promptlayer (tbh i was until recently of the opinion that you should always check in your prompts into git.
343.
by swyx 9mo ago
i'm sorry for your porcupine :(
344.
by swyx 9mo ago
say more about what you were looking for that promptlayer had/langfuse didnt?
345.
by swyx 9mo ago
Altinity isn't just a vendor, it's THE ClickHouse vendor before ClickHouse became a company. https://altinity.com/blog/big-news-in-the-clickhouse-communi... always nice to see a database ceo be "one of u
346.
by swyx 9mo ago
(congrats team! such a joy to see you succeed) every single day there is an acquisition on here. what's going on in the macro?
347.
by swyx 9mo ago
all DoS attacks and one XSS. this isnt as bad as the react server components CVEs, which enabled RCE. saving people a click: CVE-2026-22775: DoS in devalue.parse due to memory/CPU exhaustion > Effects: A malicious payload can cause
348.
by swyx 9mo ago
fyi i maintain a repo that accidentally tracks github actions cron reliability ( https://www.swyx.io/github-scraping ) - just runs a small script every hour. i just checked and in 2025 there was at least 2 outages a month eve
349.
by swyx 9mo ago
as a first step, let's tax these things. this is such an immense waste of electronics.
350.
by swyx 9mo ago
hi will! super nicely written, nice look under the hood of your processing. as an orchestration guy i always wondered why everyone seems to converge on using Ray, and as a secondary thought, how well is Anyscale capturing the Ray market. if
351.
by swyx 9mo ago
> . I've experimented with this as a metric for distinguishing whether text is written by AI. Human-written text doesn't follow a constant-temperature softmax distribution, either. oo that sounds like a cool insight. like just
352.
by swyx 9mo ago
"What might be more surprising is that even when we adjust the temperature down to 0This means that the LLM always chooses the highest probability token, which is called greedy sampling. (thus making the sampling theoretically determin
353.
by swyx 9mo ago
interesting exercise and well written. my followon questions/work would be: 1a. temperature=100000 is interesting too. obviously "ideal" temperature lies somewhere between 0 and 100000. has anyone ablated temperature vs intel
354.
by swyx 10mo ago
sharing my list of mistakes in markdown that Gruber endorsed :) https://news.ycombinator.com/item?id=22776108
355.
by swyx 10mo ago
no, they were not made to do it. they listened to feedback and did the work. this is better than we get in 99% of cases. try to be nicer and meet them half way instead of living in your ideal world.
356.
by swyx 10mo ago
the frontend is beautiful. i find it inspiring that you have 10 years of data science and are no longer limited by your lack of frontend or design knowledge. this is a better site than i couldve done
357.
by swyx 10mo ago
> Jeff & Sanjay did everything together, my nonexpert impression is jeff keeps much more of a public profile. hence the natural celebrity goes to him. was this not true way back in the day?
358.
by swyx 10mo ago
i asked them this in my interview. tldr they subsidize all inference on their platform https://www.youtube.com/watch?v=NBnOk0Uy9ig&t=70s
359.
by swyx 10mo ago
> The results I use in my dashboard are from DistilBERT because it runs efficiently in my Cloudflare-based pipeline. interesting - why use cloudflare vs say hf inference or modal? and is this replicate-cloudflare or normal cloudflare?
360.
by swyx 10mo ago
he pretty much always says it offline haha but i maay have mixed it up with the subsequent convo we had at neurips https://www.latent.space/p/neurips-2023-startups

More ›