punk 3983
gp @ 0x5f capital
ceo ethos
/ethos-network

me a few months ago: gpt-4.1 has a 1 million token context window all my problems are solved

me today: let’s bring that splitting code back out, eventually we’ll reach the limit to all space provided

Technowatermelon. Elder Millenial. Building Farcaster. 

nf.td/varun

“Fruits and vegetables keep us alive always remember to eat your five!” - Super Potato

are there good examples of media companies that have leaned into Farcaster?

Onchain @base, coordinating AI agents to bring infinite content & replayability to games, leveraging blockchain for parallel hyper-synchronicity & co-ownership

Earlier this week I articulated something that is probably worth sharing in writing. We have a situation where talented young people are entering a tight work force (for software engineers at least). They will probably end up creating a lot of new businesses for the next generation

Dialectic with Jackson Dahl

Talk: /dialectic Write: jdahl.substack.com
Before: Investor /paradigm, Founding team /100thieves
jacksondahl.com
/jackson

I had to split it into two listening sessions, but the one with Alex Danco is really really good!

spotlight

Introducing $PROCOIN, the community-led governance token for Farcaster Pro users.

All 10k NFT holders are eligible for an airdrop.

But there’s a twist.

**The token will be untransferable until 500 Pro users claim.**

In order to get a mandate as a community token, we need broad support.

To claim, quotecast this post with the text “Ratifying @procoin governance”

All 10k NFT holders are eligible, but the first 500 claimers will receive ~5x the allocation.

$PRO 0xF65C3c30dd36b508E29a538B79b21e9b9E504E6C | Probots | Procoin Curation Protocol coming soon

happy to try out pro “for the culture”, I will say that it was kind of weird not to see a paymaster for the tx. I might be an abnormal user but it would be a ux win for everyone

teddit

building farcaster | surfing @ venice, ca | eating @ gjusta / gjelina | “lucky me, lucky mud” 🤠

I’ve never had, but what makes a lemonade diet?

I was 2 km away from today's mystery location 📍. Can you beat me?

I was 0 km away from today's mystery location 📍. Can you beat me?

69ism

🍌 @revealcam | playing around onchain: @alfaca @offmarket @perl | vc warm intros | fid #539

with hours and hours of audio to process, Meridian is on the back of a lot of LLM work. Initially I wanted a tool to provide trend data and the newsletter is a tangible way for me to justify the work. Particularly because it provides value to people without having to do the work.

yesterday I simplified something that should make back search a lot easier: deepgram diarizied data -> split by speaker with metadata -> vectorized -> into pgvector. I should now be able to do give me references Cosmos within this date range, and who did the mentions etc

One of my goals for Meridian is to ensure that it feels as native to the internet as possible. This will evolve as needed to resonate with the audience of today and tomorrow. With yesterday's issue, I also published a companion playlist that you should checkout

One of my goals for Meridian is to ensure that it feels as native to the internet as possible. This will evolve as needed to resonate with the audience of today and tomorrow. With yesterday's issue, I also published a companion playlist that you should checkout https://blog.getmeridian.io/p/playlist-eth-scaling-solutions-by-2026-btc-edges-toward-83k

I asked claude 3.7 Sonnet to write me a prompt for gpt-4o for the speaker assignment and now I have something I can pass a transcript to ,it samples, and provides predictions of which speakers are which

with that said, feel free to sign up to https://www.getmeridian.io/

We sort of do this heuristic in our head of eliminating who the speaker isn't each time we listen to a new podcast. Since we can give the model a fair bit of context it should be able to knock it all down

The last issue of the newsletter covered 22 podcasts, that is a pretty heavy time investment for each issue. So I decided to see how far I could get with some of the easily accessible models

Akeem

@akeem