HNHacker News
TopNewBestAskShowJobs

swyx

24,625 karma · joined December 9, 2015

Writer/Angel in devtools and smol AI enjoyer.

ai engineer pod/news: https://latent.space/

conference: https://ai.engineer

devrel/devtools: https://dx.tips/

personal blog: https://swyx.io/ideas

book: https://learninpublic.org/

but you can also email username [at] ai dot engineer

---

meet.hn/city/37.7792588,-122.4193286/San-Francisco

Socials:

- x.com/swyx - youtube.com/@swyxtv

submissionscomments
swyx··on Claude Code daily benchmarks for degradation tracking
chill out, ofir does not work for anthropic. he's just saying there's inherent variability in LLMs and you need to at least 30x the samples that OP is doing in order to make any form of statistically significant conclusions.
swyx··on Claude Code daily benchmarks for degradation tracking
they could but you can also have some trust in anthropic to have some integrity there, these are earnest people.

"trust but verify" ofc . https://latent.space/p/artificialanalysis do api keys but also mystery shopper checks

swyx··on Prism
not at all about stealing karma, i dont care much about fake internet points.

yes you got the important thing!

swyx··on Pandas 3.0
yes but a lot of legacy code wil still need to be maintained and written. you dont see how this can be confusing/annoying?
swyx··on Pandas 3.0
its not even about the ingest, every major semver change now is a problem because now LLMs will need to contextually distinguish whether or not they are expected to output Pandas 2 or 3, unless ofc you explicitly prompt it.
swyx··on Show HN: LemonSlice – Upgrade your voice agents to real-time video
this is like Tavus but it doesnt suck. congrats!
swyx··on Prism
oh i was here to post it haha - thank you for doing that job for me so I'm not a total shill. I really enjoyed meeting them and was impressed by the sheer ambition of the AI for Science effort at OAI - in some sense I'm making a 10000x smaller scale bet than OAI on AI for Science "taking off" this year with the upcoming dedicated Latent Space Science pod.

generally think that there's a lot of fertile ground for smart generalist engineers to make a ton of progress here this year + it will probably be extremely financially + personally rewarding, so I broadly want to create a dedicated pod to highlight opportunities available for people who don't traditionally think of themselves as "in science" to cross over into the "ai for hard STEM" because it turns out that 1) they need you 2) you can fill in what you don't know 3) it will be impactful/challenging/rewarding 4) we've exhausted common knowledge frontiers and benchmarks anyway so the only* people left working on civilization-impacting/change-history-forever hard problems are basically at this frontier

*conscious exaggeration sorry

swyx··on Prism
we did a podcast with the Crixet founder and Kevin Weil of OAI on the process: https://www.youtube.com/watch?v=W2cBTVr8nxU&pp=2Aa0Bg%3D%3D
swyx··on Jurassic Park - Tablet device on Nedry's desk? (2012)
ok so it now begs the question... whos plane was this?
swyx··on Show HN: Mastra 1.0, open-source JavaScript agent framework from the Gatsby devs
use mastra, get acquired!
swyx··on Capital One to acquire Brex for $5.15B
sorry how did employees get fucked? theres more money after the 1.7B.
swyx··on Capital One to acquire Brex for $5.15B
he actually did not! came together quickly. https://x.com/jamesreggio/status/2014520219847557619?s=20
swyx··on Capital One to acquire Brex for $5.15B
Brex's CTO recently came on LS to talk about their AI strategy and tech: https://latent.space/p/brex
swyx··on Devin Review: AI to Stop Slop
hahahah oh no
swyx··on Show HN: Mastra 1.0, open-source JavaScript agent framework from the Gatsby devs
ty! btw your docs search is broken, as i tweeted to Ward.
swyx··on Show HN: Mastra 1.0, open-source JavaScript agent framework from the Gatsby devs
> a `.network()` method that turns any agent into a routing agent

say more pls?

swyx··on Show HN: Mastra 1.0, open-source JavaScript agent framework from the Gatsby devs
> 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!!!

for those who want an independent third party endorsement, here's Brex CTO talking about Mastra in their AI engineering stack http://latent.space/p/brex

swyx··on I was a top 0.01% Cursor user, then switched to Claude Code 2.0
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 every project”

“sick, what are you building?”

“my setup is so optimized, i’m using like 5 instances at once”

https://x.com/johnpalmer/status/2012911338276720852?s=46

swyx··on X For You Feed Algorithm
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 shared without weights so we only have so much visibility into the actual influence of each trait.

"We have eliminated every single hand-engineered feature and most heuristics from the system. The Grok-based transformer does all the heavy lifting by understanding your engagement history (what you liked, replied to, shared, etc.) and using that to determine what content is relevant to you." aka it's a black box now.

the README is actually pretty nice, would recommend reading this. it doesnt look too different form Elon's original code review tweet/picture https://x.com/elonmusk/status/1593899029531803649?lang=en

sharing additonal notes while diving through the source: https://deepwiki.com/xai-org/x-algorithm

and a codemap of the signal generation pipeline: https://deepwiki.com/search/make-a-map-of-all-the-signals_3d...

- Phoenix (out of network) ranker seems to have all the interesting predictive ML work. it estimates P(favorite), P(reply), P(repost), P(quote), P(click), P(video_view), P(share), P(follow_author), P(not_interested), P(block_author), P(mute_author), P(report) independently and then the `WeightedScorer` combines them using configurable weights. there's an extra DiversityScore and OONScore to add some adjustments but again dont know the weights https://deepwiki.com/xai-org/x-algorithm/4.1-phoenix-candida... - other scores of interest: photo_expand_score, and dwell_score and dwell_time. share via copy, share, and share via dm are all obviously "super like" buttons.

- Two-Tower retrieval uses dot product similarity between user features/engagement (User Tower) and normalized embeddings for all items (Candidate Tower). but when you look into the code and considering that this is probably the most important model for recommendations quality.... it's maybe a little disappointing that its a 2 layer MLP? https://deepwiki.com/search/what-models-are-used-for-user_98...

- Grok-1 JAX transformer (https://github.com/xai-org/x-algorithm/blob/main/phoenix/REA...) uses special attention masking that prevents candidates from attending to each other during inference. Each candidate only attends to the user context (engagement history). This ensures a candidate's score is independent of which other candidates are in the batch, enabling score consistency and caching. nice image here https://github.com/xai-org/x-algorithm/blob/main/phoenix/REA...

- kind of nice usage of Rust traits to create a type safe data pipeline. look at this beautiful flow chart https://deepwiki.com/xai-org/x-algorithm/3-candidate-pipelin... and the "Field Ownership pattern" https://deepwiki.com/xai-org/x-algorithm/3.6-scorer-trait#fi...

- the ten pre-scoring filters are minorly interesting, nothing super surprising here apart from AgeFilter (https://deepwiki.com/xai-org/x-algorithm/4.6.1-agefilter) which I guess means beyond a certain max_age (1 day?) nothing ever shows up on For You. surprising to have a simple flat cutoff vs i guess the alternative of an exponential aging algorithm.

- videoduration hydrator explicitly prioritizes video duration (https://deepwiki.com/xai-org/x-algorithm/4.5.6-videoduration...) but we dont know in what direction... do you recommend shorter or longer videos? and why a hydrator for what is presumably a pretty static property?

open questions from me

1. how large is the production reranker? default param count is here https://deepwiki.com/search/how-many-params-is-the-transfo_c... but that gives no indication. the latency felt ultra high initially last year and seems to have come down some, what budget are we working with?

2. can we make the retrieval better? i dont have a tooon of confidence in the User Tower / Candidate Tower system - is this SOTA (it's probably not - see how youtube does codebook semantic id's https://www.youtube.com/watch?v=LxQsQ3vZDqo&list=PLcfpQ4tk2k... )

3. no a/b testing / rollout infrastructure?

4. so many hydration subsystems - is this brittle?

swyx··on Nanolang: A tiny experimental language designed to be targeted by coding LLMs
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:

Minimal syntax (18 keywords vs 32 in C)

One obvious way to do things

Tests are part of the language, not an afterthought

Transpile to C for maximum compatibility

ehh. i dont think the overhead of inventing a new language makes up for the lack of data around it. in fact if you're close enough to rust/c then llms are MORE likely to make up stuff from their training data and screw up your minimal language.

(pls argue against this, i want to be proven wrong)

swyx··on /R/selfhosted limits vibecoded apps
> HN is obviously very pro-ai

mm i'd actually say its more moderated, lots of cynics/skeptics.

swyx··on I set all 376 Vim options and I'm still a fool
maybe throw it in your friendly neighborhood LLM
swyx··on I set all 376 Vim options and I'm still a fool
> single keystroke could move the cursor halfway across the file to exactly the right spot.

sorry what is this? exaggeration?

swyx··on Dead Internet Theory
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
swyx··on A Social Filesystem
> 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?

swyx··on A Social Filesystem
> 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 useful if partially corrupted), but economic incentives keep pushing us to keep things in code (brittle, dies basically when one of maintainer|buildtools|hardware substrate dies).

when standards emerge (forcing code to accept/emit data) that is worth so much to a civilization. a developer ecosystem tipping the incentive scales such that companies like the Googl/Msft/OpenAI/Anthropics of the world WANT to contribute/participate in data standards rather than keep things proprietary is one of the most powerful levers we as a developer community collectively hold.

(At the same time we shoudl also watch out for companies extending/embracing/extinguishing standards... although honestly outside of Chrome I struggle to think of a truly successful example)

swyx··on ClickHouse acquires Langfuse
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... flipped recently)
swyx··on Cursor's latest “browser experiment” implied success without evidence
i'm sorry for your porcupine :(
swyx··on ClickHouse acquires Langfuse
say more about what you were looking for that promptlayer had/langfuse didnt?
swyx··on ClickHouse acquires Langfuse
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 us" and/or "write like a real human being".

← PreviousPage 12 of 34Next →