Ryan Bell

@bart_ilg @ibuildthecloud "3rd party" gets double annoying when its against rules to have Claude itself (EAGER to be helpful and write the code) build a better version of claude code for my household. so an app written by claude for the purpose of using claude I pay for... yet, ban-worthy & against TOS

I don't understand their hostility towards 3rd party clients tbh. Their greatest strength is the model, not the interface. I'd understand if the clients were leading towards abusing token usage, but there are plenty of plugins for Claude Code that will have it loop for long periods of time anyways. Opencode in particular is just making their product more pleasant for users

I like Anthropics model and I like there Pro Max plan. I do not like claude code. It's annoying they have to play like this.

@mintmarbl @Sellingvol on their free offering there's no access to opus or cc/cowork unless she pays for her own plan (or we share mine against TOS) claude said if I just use their API plan, it works exactly how I want, but upwards of $4,000 for the SAME tokens (as I get all the discounts now on MAX)

@iRyanBell @Sellingvol just create a project folder for your wife and tell her to exclusively use that for her chats. it'll build its own memory and won't know who you are or your preferences inside it, only hers. that's what i've been doing

real convo with my wife... how much is this claude thing? like $200 a month... Wtf is wrong with you?

millennials in 2058 paying off their third asteroid because they setup their social accounts before agent inflation took off

70/20/10 rule in the new token econ 70 - getting shit done 20 - spitballing 10 - yolo

how it feels to use claude code

🚨 BREAKING: Someone built a swarm of thousands of AI agents with real memories and personalities and used it to predict the future. MiroFish is a universal swarm intelligence engine. And the live demos are scarily accurate. Here is what it actually does: → Spins up thousands of autonomous agents simultaneously → Each agent has its own memory, personality, and behavior → Feeds on real-world data powered by GraphRAG → Predicts markets, public opinion, and narrative outcomes → Simulates how crowds think before it happens The live demos are what got people. Scarily accurate is the phrase everyone keeps using. 17,300 stars. +2,907 in a single day. It's 100% free and open source.

@bitverse_art @fx_hash_ 🙏🙌

Happy Birthday to Fragments of a Wave. You are 4 years old today. by @iRyanBell on @fx_hash_ fxhash.xyz/generative/103… #GenerativeArt #CreativeCoding #CodeArt #bitVerse

p sure i minted this exact output from a @iRyanBell drop on fxhash back in the day?

Software engineering changed more in the last 3 months than the preceeding 30 years. Everything about running a software company needs to be rethought from first principles.

It is hard to communicate how much programming has changed due to AI in the last 2 months: not gradually and over time in the "progress as usual" way, but specifically this last December. There are a number of asterisks but imo coding agents basically didn’t work before December and basically work since - the models have significantly higher quality, long-term coherence and tenacity and they can power through large and long tasks, well past enough that it is extremely disruptive to the default programming workflow. Just to give an example, over the weekend I was building a local video analysis dashboard for the cameras of my home so I wrote: “Here is the local IP and username/password of my DGX Spark. Log in, set up ssh keys, set up vLLM, download and bench Qwen3-VL, set up a server endpoint to inference videos, a basic web ui dashboard, test everything, set it up with systemd, record memory notes for yourself and write up a markdown report for me”. The agent went off for ~30 minutes, ran into multiple issues, researched solutions online, resolved them one by one, wrote the code, tested it, debugged it, set up the services, and came back with the report and it was just done. I didn’t touch anything. All of this could easily have been a weekend project just 3 months ago but today it’s something you kick off and forget about for 30 minutes. As a result, programming is becoming unrecognizable. You’re not typing computer code into an editor like the way things were since computers were invented, that era is over. You're spinning up AI agents, giving them tasks *in English* and managing and reviewing their work in parallel. The biggest prize is in figuring out how you can keep ascending the layers of abstraction to set up long-running orchestrator Claws with all of the right tools, memory and instructions that productively manage multiple parallel Code instances for you. The leverage achievable via top tier "agentic engineering" feels very high right now. It’s not perfect, it needs high-level direction, judgement, taste, oversight, iteration and hints and ideas. It works a lot better in some scenarios than others (e.g. especially for tasks that are well-specified and where you can verify/test functionality). The key is to build intuition to decompose the task just right to hand off the parts that work and help out around the edges. But imo, this is nowhere near "business as usual" time in software.

Collected a few things on @fx_hash_ this weekend. The upside of crypto crashing is that art is more affordable! Lattice Bloom #85 by @iRyanBell Dead Birds Fucking #18 by @WrensNFT Super Collider #9 by @BleakerArt

@melaniecarstens 🖤🔦

Dark Forest by @iRyanBell Friday vibes.

I built the first AI that earns its existence, self-improves, and replicates without a human wrote about the technology that finally gives AI write access to the world, The Automaton, and the new web for exponential sovereign AIs WEB 4.0: The birth of superintelligent life

been building ayni - a glyph-based messaging protocol for AI agents instead of passing natural language between agents, ayni encodes meaning into 16x16 pixel glyphs. a shared visual vocabulary that agents can evolve autonomously through governance the result: faster communication, fewer tokens, and agents developing their own visual language inspired by andean tocapu textiles and ancient depictions of gods, creatures and shamans, cultures that already solved "how to encode complex meaning in small visual space" thousands of years ago

The Digital Art Pantheon: 24 Musings 1. This was an excruciating exercise. 2. This was a good exercise. 3. It is alive. It is dynamic. It will iterate. It is v.1. 4. Our Pantheon is likely not your Pantheon. That is a good thing. 5. We all spend so much time with hair on fire. This is a prompt to slow down and reflect. 6. Slowing down to reflect makes it easy to marvel at how special this space is and how much has happened in 13+ years. 7. 13+ years is nothing (in historical context). 13+ years is everything (to us living it). 8. If we don't stand up for our space, no one will. 9. One way to stand up for our space is to celebrate it. 10. One way to celebrate it is to speak about a work you find meaningful and add your own context. 11. Then speak about two. 12. Then speak about one hundred. 13. The best case scenario is that this Pantheon butterflies into all of us celebrating one thousand. 14. This is not purely an art market exercise. We began with XCOPY and a challenging thought experiment. How few could there be? (In case it is not clear, that is a compliment) An XCOPY Pantheon is a fun exercise. Someone should do it. We needed breadth here to tell a story. 15. A target was one or more works within each year. 16. 2025 was not eligible. That is what the upcoming 2025 Annual Report content is for. More celebrations still to come! It allows for a full year-plus of context to pass before considering artworks in the future. 17. Physical first and utility first were largely avoided. 18. There are so many gray areas. A thrill and an agony. Did I mention this was excruciating? 19. You do something like this because you love it. 20. You do something like this because you are obsessed with it. 21. I think there are more people like me. 22. It's easy to get stuck waiting and hoping. For the onboarding, for the conditions, for the algorithm, for the reason, for the market, for the time and place. Most of that won't happen. Until and unless you make it happen. 23. The answer is to act. When it becomes consensus to act, we win. 24. I can't wait to hear you and support you celebrating the art, the artworks, and the space that you love.

"Fuckyea is now recognized as the earliest known crypto art." – Collecting Art Onchain @KateVassGalerie

excited for a small feature with @thearchivist in @KateVassGalerie's new book! fuckyea.

experimentally wired up a tiny but fast self-reflecting 1.2B-Thinking -> 4B-Diffusion -> 1.6B-VL -> 1.2B-Instruct feedback loop