Vibe Coding That Works!

It's the dream! This has been the culmination of work over several months now, but I'm now fairly happy that I've cracked a workable methodology for coding rapidly with AI and I sat down with Andy today to discuss on the Good Stuff Podcast.

(Links below - fountain will follow once it all syncs - I really need to setup a Haven blossom server...)

We call it "Slow Code",

Slow Code, like Slow Food, not because it's slow but because its deliberate and intentional.

Going to be posting more about this, but I haven't been this excited since I heard nostr:nprofile1qy0hwumn8ghj7cnfw33k76twd4shs6tdv9kxjum5wvhx7mnvd9hx2qg4waehxw309ajkgetw9ehx7um5wghxcctwvsqzq3e0gs8jnmued6f2rp4c6vs07xqvs4vs8zpwt82smcdch4txjvq76kl2yj say the words "Sound Money" and I thought "that sounds better than not sound money - I should look into it".

YOLO mode

No-Stir -> black coffee

That explains it

nostr:nevent1qqsqxr0sgleftdys85mgnpj59v6d2682up4zdu33xh4sz9eulgz609cfmkngp

Shitcoining wasn’t shitcoining because of the coin, it was because of the shit.

We don’t have to forget that just because sometimes it’s painted orange.

For ideology it does matter who controls your system prompt?

As a quick experiment over coffee this morning I took that last episode of nostr:nprofile1qyxhwumn8ghj7mn0wvhxcmmvqy0hwumn8ghj7mn0wd68ytn9d9h82mny0fmkzmn6d9njuumsv93k2qpqn00yy9y3704drtpph5wszen64w287nquftkcwcjv7gnnkpk2q54svljnn3 , generated a transcript and ran it through a dialogue agent that runs a 10 round discussion between pre-programmed AI agents around a given prompt.

In this case it was "What is the Underlying Philosophy of Sovereign Engineering based on this discussion between nostr:nprofile1qyxhwumn8ghj7e3h0ghxjme0qyd8wumn8ghj7urewfsk66ty9enxjct5dfskvtnrdakj7qpql2vyh47mk2p0qlsku7hg0vn29faehy9hy34ygaclpn66ukqp3afqta478g and nostr:nprofile1qy28wue69uhnzv3h9cczuvpwxyargwpk8yhsz3rhwvaz7tmed3c8qarfxaj8s6mrw96kvef5dve8wdrsvve8vvehwamxx7rnwejnw6n0d3axu6t3w93kg7tfwechqutvv5ekc6ty9ehku6t0dchsqgrwg6zz9hahfftnsup23q3mnv5pdz46hpj4l2ktdpfu6rhpthhwjv0us2s2 "

Usually I run most things on Claude 4 Sonnet as it s great general purpose model, but i was surprised at how it seemed to crowbar into the conversation a bunch of AI safety viewpoints and ideologies that aren't necessarily in the source transcript.

So I figured lets change out the model for the dialogue agents to Grok-4 and see what it does instead - does the ideology leak out still? Is it different?

I actually think that Grok 4's first assessment kinda nailed it:

> Sovereign Engineering seems like a mindset or movement for building tech in a way that's deeply human-centered, decentralized, and empowering—think vibing with AI and open protocols like Nostr to create tools that foster freedom, collaboration, and realness without the chains of big tech overlords.

Full summaries from both agents and transcripts are available here:

https://github.com/humansinstitute/everest-pipeliner/wiki/Two-AIs-Analyse-No-Solutions-to-extract-the-Philosophy-of-Sovereign-Engineering

Interesting to see how repeated calls back to agents can lead down a specific pre-programmed ideological path. I think I prefer models with ideologies close to my own, but the alternative is a good echo bubble popper 😉

My guess here is this is more driven by the system prompt / safety layer than the training.

Its a great read.

It was a really solid moment whrn nostr:nprofile1qy0hwumn8ghj7cnfw33k76twd4shs6tdv9kxjum5wvhx7mnvd9hx2qg4waehxw309ajkgetw9ehx7um5wghxcctwvsqzq3e0gs8jnmued6f2rp4c6vs07xqvs4vs8zpwt82smcdch4txjvq76kl2yj brings up The Mandibles to nostr:nprofile1qy2hwumn8ghj7etyv4hzumn0wd68ytnvv9hxgqgdwaehxw309ahx7uewd3hkcqpqxeejes6lu4scttcmzytq5wfad3e6rljpmhc3snqs89xz3jjavfasml4lsz - Balaji jumps on "great book" and you can hear Marty shrinks audibly fearing for follow ups

You can almost hear nostr:nprofile1qyv8wumn8ghj7urjv4kkjatd9ec8y6tdv9kzumn9wsq3vamnwvaz7tmjv4kxz7fwwpexjmtpdshxuet5qqsqfjg4mth7uwp307nng3z2em3ep2pxnljczzezg8j7dhf58ha7ejgqgzx3h enter the chat

Peak performance

You mean the patented Human At The Edge methodology 🤣

I think I need to just get a mac mini running a bunch of VMs in Yolo mode 24x7

https://primal.net/e/nevent1qvzqqqqqqypzp9pptap2jce4epluh85grg9mkc4e5724r8ssnn6ln5h0v9mgra3zqyt8wumn8ghj7etyv4hzumn0wd68ytnvv9hxgtcpr9mhxue69uhhqun9d45h2mfwwpexjmtpdshxuet59uq32amnwvaz7tmjv4kxz7fwv3sk6atn9e5k7tcqyqvckmuggls6cw4dy4ngsnafxgxywesxxscrdctc5wu3dze379pqgx649hx

lol -> https://www.otherstuff.studio/thegoodstuff

You are not stalking me hard enough.

Posting transcripts and summaries from a couple of AI dialogues, that take a podcast and an idea to explore then work through it in a dialogue.

If you like that check out the source material in in the podcast ;)

https://github.com/humansinstitute/everest-pipeliner/wiki

nostr:nevent1qvzqqqqqqypzp9pptap2jce4epluh85grg9mkc4e5724r8ssnn6ln5h0v9mgra3zqyt8wumn8ghj7etyv4hzumn0wd68ytnvv9hxgtcpr9mhxue69uhhqun9d45h2mfwwpexjmtpdshxuet59uq32amnwvaz7tmjv4kxz7fwv3sk6atn9e5k7tcqypsa3xd0gfl4u0gx94mansd484l855j2xdxcsgw5z968u26cxjm5ut52e09

The back and forth is pretty interesting. Just experimented with adding a facilitator to the group to pull them back when they go off on one, but needs tweaking.

I can post a few of the conversations if you're (anyone) is interested.

Claude went a bit straight socialist on the last one so I think I may need to add personality adjustments to the setup 🤣

Built a weekend project to allow a couple of AI Agents to discuss a subject in a dialogue back and forth to see if they develop any new ideas.

Have been feeding it my podcast transcript and using it to expand on ideas which has been very fruitful - some great ideas coming out.

I'm fairly convinced that predictability in implementation will trump raw speed in vibe coding.

Speeding yourself up 8x and being able to maintain or continue building on something complex seems way more preferable than moving 10x and having to continually throw everything away

Its all about "Super Fast Waterfall" althoguh I think I prefer the term "slow code"as. methodology

The hardest thing about working with AIs is slowing them down and controlling them enough so you can actually focus and build something useful.

Sunset was solid to

nostr:nevent1qqsduwdur2nr3nmu0rwft002fxm94gktsqzwt64rsxsue9ss629r4espzemhxue69uhhyetvv9ujuurjd9kkzmpwdejhgt44pzd

“Why did you want that table” 👀

I started putting our podcast transcripts into a dialogue workflow of AI agents picking out one idea and then work shopping on it, back and forth to take ideas expressed in the podcast and develop them.

It is shockingly good at this and they've just arranged to meet up for 3 hours on Saturday to start running experiments 🤣 👀

Interesting talk on working with and promoting AIs for creative work.

Didn’t expect to go down a “What can we learn from the Shamans” route but there we go :)

https://m.youtube.com/watch?v=J_w4wxex-yM

The real 58k gang

If it’s “critical market infrastructure” it’s not “market infrastructure”

This is more of the care of data that just doesn’t need to be public.

Mongo By Relays!

It's cool that Nostr apps can be thin clients using Nostr as a free backend 😉

Sometimes though I just want a db to store some data that doesn't live forever on public relays where a boring offline DB would be sweet.

Quick example project to implement NostrMQ as a simple access layer for MongoDB requests.

-> No hosting, no api layer ephemeral keys for misdirection and encrypted responses.

- https://github.com/humansinstitute/nostrMQ/blob/main/docs/mongobyrelay.md

- https://github.com/humansinstitute/nostrMQ/blob/main/examples/mongo-pilot-client.js

Learn to build.

Don’t learn to syntax

Talk about it every week here :)

https://fountain.fm/show/5c1jIyQLUHKkpS1r0ssb

I’m fairly convinced we need highly opinionated AI agents for good UX

Have any journalists checked their signal invites for the alpha?

With the humans

Classic shirt

nostr:nprofile1qy28wue69uhnzv3h9cczuvpwxyargwpk8yhsz3rhwvaz7tmed3c8qarfxaj8s6mrw96kvef5dve8wdrsvve8vvehwamxx7rnwejnw6n0d3axu6t3w93kg7tfwechqutvv5ekc6ty9ehku6t0dchsqgrwg6zz9hahfftnsup23q3mnv5pdz46hpj4l2ktdpfu6rhpthhwjv0us2s2 nostr:nprofile1qyxhwumn8ghj7e3h0ghxjme0qyd8wumn8ghj7urewfsk66ty9enxjct5dfskvtnrdakj7qpql2vyh47mk2p0qlsku7hg0vn29faehy9hy34ygaclpn66ukqp3afqta478g

Are you running Sovereign Engineering next year?

Wife heard me talking about it as a bucket list thing to do on a podcast and now i need to make a case for it 😉

Esperance is lovely nostr:npub1jqckepsld3xn98aeq7yg72g0yrqkz92vegkv6k3prfhkzu356v5qa6akee

nevent1qqsqksrsvx67w8mes63c7frtz0wkddppt48meefdaa3h27lgdltrrwcpupmhxue69uhhyetvv9ujuerpd46hxtnfduhj2v3swaehxw309aex2mrp0yhxumm5daeks6fwwa5kute9xgc8wumn8ghj7mn0wvhxcmmv9ujnyvrhwden5te0wfjkccte9eekjctdwd68ytnrdakj7ffjxpmhxue69uhhyetvv9ujuvrcvd5xzapwvdhk6te9xgc8wumn8ghj7mnxwfjkccte9eshqup0y5erqamnwvaz7tmjv4kxz7tjwvhxumm5daeks6fwwa5kute9xgc8wumn8ghj7un9d3shjtnwv4u8getj0ghxxmmd9ujnyvrhwden5te0vejkuunfwgkhxtnwda6x7umgdyh8w6twd9dxj3

Skipping 3.7 was probably just a wise move!

I imagine they'll have it live by the end of the day, seems to be the way!

I swear if there is a bug in this app I'm screwed though its been running unattended for a while now ;)

Just watching Claude 4 eating all of my fresh credits 😉

But there is something nice about just yolo-vibing an app from nothing to see if it works ;)

Sat down with Princey to talk AI and Bitcoin :)

https://fountain.fm/episode/0WVKg8jWoLBIufiD8tic

Nice :)

I already have a can parked outside the front door with everything all ready to go which solves the disaster problem :)

Sorry to hear about the tornado!

Sometimes I wonder about just wiring a secondary 12v only circuit into the house and running everything off a couple of camp batteries

Bar the oven (I could live off the BBQ) and fridge I think everything else I actively step down from 240v, seems like it could be simpler.

To be fair the last time nostr:nprofile1qqsgydql3q4ka27d9wnlrmus4tvkrnc8ftc4h8h5fgyln54gl0a7dgspz3mhxue69uhhyetvv9ujuerpd46hxtnfduq32amnwvaz7tmjv4kxz7fwd4hhxarj9ec82csafn79u used sats he was accused of being a Russian double agent.

I’m expecting a bit of “security budget”next

The Good Stuff 01 - The business model of AI

https://podcastindex.org/podcast/7297956?episode=36136586157

All doors, could be secret hidden bookcase doors.

Jan 03 - Proof of Keys Game

This year I set the kids a challenge to teach key management :)

1. We create a brand new bitcoin seed each (12 words)

2. We deposit $365 of bitcoin to each seed ($1 a day)

3. Each kids has to secure these 12 words for 1 year from their parents.

4. If they can recover the wallet in 1 year they get the bitcoin as spending money.

5. If we (or anyone else!) discover it bitcoin gone, lessons learned.

Will do check ins every 3 months I think and I'll slowly add in various approaches to how to secure things as we go.

Oldest is 10 so this is serious money to them :)

They are very excited for this :)

#bitcoin #ProofofKeys