building phrasing.app with @elixirlang & #clojurescript if you're an engineer and polyglot dm me 📥 let's go duo hunting

Amsterdam
Based in France
Starting to look like a set. @figma has been busy.
9
13
734
If you want to do it all on your own, you can. If you want to do all you can do, you can’t on your own.
Awesome founder with an awesome product. Check him out on PH!
Launching DoneThat on ProductHunt today! DoneThat enables more remote work by tracking work in useful and privacy-sensitive chunks. producthunt.com/products/don…
1
LLMs are the electron-app-ification of machine learning
Too short. Too long. Just right. Your expressions shouldn't need a caliper to measure. But now one exists.
1
@figma said anything is possible. I took that personally and built a caliper.
1
Apologies on behalf of English
1
Been using @MistralAI large 4 a lot via it’s chat UI. I’m very impressed. It’s been good enough to replace all my other chat UI’s. It’s omniscience isn’t good enough to replace frontier models in phrasing, but it’s great for daily tasks
1
Launch week one. Everything that shipped, why it shipped, and how it was built.

phrasing.app/blog/launch-wee…
1
2
Three years building. A year launching. Always crickets. Here's how I keep going anyway. phrasing.app/blog/how-do-you…
3
Where it all started. Dark mode, command mode, and WebGL analytics shipped in one week back in January. phrasing.app/blog/launch-wee…
1
Started with 10 languages. Now it's 20. The timeline is fine. Everything is fine. Language Log 006 out now.

piped.video/ss6zf9RMVAo
Replying to @jeremyaube_
Absolutely the right direction, but not a big enough model. Hoping they catch on more as an alternative to reasoning models! Although terrified of the resulting product naming disaster that would follow 😂 gpt-io gpt-jupiter and… gpt-andromeda?
1
1. No thinking normally does not improve the accuracy. I’ll try the new models but so far thinking actually increases the error rate of basic questions 2. Yes I know POS taggers exist. In practice LLMs outperform spacy/stanza/nltk
Am I the only one who doesn’t want reasoning with LLMs? Almost all modern LLMs now require reasoning. I don’t need my LLM to solve physics for me. I just need it to tell me if a word is a noun and be wrong less and less.
2
2 months ago Grok told me it would take me 2 years to do the full stack on pec deck. Grok was wrong 😈
1
I feel so blessed 🥹 everyone is so inspired by my site!
Immersion without spaced repetition is decoration with no hooks. Nothing sticks. phrasing.app/blog/hooks
1
Replying to @toituhoc1102
Exactly. And most people only do one or the other.
Comprehensible input is cardio. Spaced repetition is the gym. Traditional learning is a HIIT class. New episode on why that analogy holds up.

piped.video/s3W1Yb_sRsQ
1