SAM COUCH

@nickgraynews @dennishegstad @vercel +1 for cloudflare with astro.build is super easy and cheap

@dennishegstad @vercel is this a serious question!! why are you not considering Cloudflare?? hosting is basically free and it will be faster than anything else

Where would you deploy and host a blog? I can't justify the @vercel usage costs for a simple blog I was testing some AI stuff with. $50+ a month in usage costs to deploy a blog without a CMS or any back-end is not interesting. Not a fan of webflow anymore... Where are you hosting blogs with +1000s of pages?

@ssigsbee New site looks great!

Imawhale is going all-in on prediction markets. $500M+ in volume #1 on Kalshi volume leaderboard (opt ins) This is just the starting point. We’re building the leading PM trading platform. Strategic partners + hires: imawhale.com

MaaS™️- Markdown as a service

@ugmonk Ready for an upgrade!

Big things coming. A year in the making... Friday 10:00am EDT 👀

@ryancarson Do you have any special sauce to your symphony setup? I tried it for a few days and it just seemed very token hungry and took much longer than other workflows to get the same things done

Okay, explored the codebase more, and this is not a replacement for Symphony. It lacks the orchestration ability to shepherd issues from to-do to merged. I think that final version is something like Open SWE + Symphony.

Seriously. Does Twilio have any Devrel anymore?

@MediaKing MarketBeat NFT

Podcast tour is over (almost). Consulting calls are wrapped up. Private jet is at the shop for repairs. Watch collection is in the safe. Back to figuring out how to grow MarketBeat from 50m/year to 75m/year.

Ruh roh 😏

@tobi What’s your pi setup look like? What kind of niceties do you have?

OK this thing is totally insane. Before going to bed I... * used try to make a new qmdresearcher directory * told my pi to read this github repo and make a version of that for the qmd query-expansion model with the goal of highest quality score and speed. Get training data from tobi/qmd github. * woke up to +19% score on a 0.8b model (higher than previous 1.6b) after 8 hours and 37 experiments. I'm not a ML researcher of course. I'm sure way more sophisticated stuff is being done by real researchers. But its mesmerizing to just read it reasoning its way through the experiments. I learned more from that than months of following ml researchers. I just asked it to also make a new reranker and its already got higher base than the previous one. Incredible.

@MediaKing What % opt in to all 4?

Most newsletter operators collect 1 opt-in at signup. We collect 4: • Primary email list • Second email list (separate ESP) • SMS number • Browser push notifications One subscriber. Four channels to reach them. Each touchpoint is another monetization opportunity.

@saradietschy Isn't this (kind of) possible with @base? cc @jessepollak

I wish I could like invest in YouTube videos I saw a rooftop dj set at 100k views and was like oh this is incredible this is gonna do numbers.. has 24 million views now lol

@bengarvey I rememeber when they're devrel party trick was to sign up for a new phone number, write a few lines of code, then have the whole audience text the number in like 2 minutes flat

@bengarvey And seemingly no indicator of the status of whatever forms I submitted to be allowed to send a text message, also seemingly no way to test in development to my own phone number or anything!

I really miss when twilio was easy to use.

@adamwathan 👀

Planning to send out the first batch of invites for ui.sh tomorrow 💪🏻

@joshpuckett Yayoi Kusama

I’m trying to be more intentional this year with broader references outside of just software. February was a deep dive into the life and work of Virgil Abloh. What do you recommend for March?

@ShakePryzby1 Which data provider have you found to be the easiest/most timely?

For 8 years I have been a slave to eSignal ($3k/yr) because of their proprietary Zanger Volume Ratio tool. What it does is it takes mid-day volume and extrapolates to get end of day readings so you can tell early on whether there's substantial volume or not - i.e. if I'm looking for a breakout buy, I want to see >150% avg volume in that first hour. I called eSignal in 2019 and asked for access to the code. They laughed me off the phone. Like, legit belly laughed. I felt so dumb. 😂😂 I have worked with 3 different software engineers trying to recreate the code over the years. Cost me many G's in the process. The tricky part was always the fact that volume throughout the day is U-shaped and not a static number, but constantly dynamic. Not a simple equation. Enter: Claude. Recreated that shit in like an hour last night. Maybe even better. Added scanning aspects. Working on some sector aspects. 10x the tool I've been using for 8 years created in a couple hours one night. I am still trying to wrap my head around the fact we all became expert software engineers overnight 🤯 Every big trader I know is going nuts creating their own software right now. Just insane the tools we have at our beck and call these days. I'll never forget Marty Schwartz calculating his own moving averages every night just a few decades ago in Pit Bull.

@codyschneiderxx Generator

I just had Claude Code build me a Facebook ad generator that can make 100+ on-brand ad variations in minutes for $0. And I made a full Notion document guide for you. It includes: 1. How to use Claude to find the pain points and desired outcomes of your ICP 2. How to use these pain points and outcomes to write ad copy variations 3. How to build a Facebook ad template entirely with code (just like the ones you see) 4. How focus Claude Code’s design so the ad feels “on-brand” 5. How to export the Facebook ads as PNGs in a zip file 6. How to bulk upload them to a Facebook ad set 7. How to use an AI data analyst to track the success of these ads Everything above is just API calls and Claude Code doing the work for you. You just come up with the ideas and polish the outputs. Like and comment "generator" and I'll send the Notion document to you

@bp22 @ComaUnitedRoad What language and/or frameworks are you mostly using?

Machine code in college. Virtually zero. If you consider excel a language maybe skill level of C- ,10 years ago. The ability to execute an idea " that would be cool if I had 20 programmers working 6 months on it" to " just go do it" is wild. Things I never heard of that I routinely use: Ubuntu Railway Postgress I also now find code by karpathy and modify it to fit my needs from GitHub. The biggest pain in vibe coding is password management. The rest is fun!!

The urge to code is overwhelming now. This vibe code thing may be a curse.

My GitHub contributions are up 944% over the last 6 months areyougoingexponential.vercel.app/samuelcouch 👀