
building https://teapawt.net/t.co/uoq0KPfOy2 / prev. game physics @Microsoft
Belgrade, Serbia
Joined November 2013
- Tweets443
- Following221
- Followers196
- Likes8.5K
I gave Fable 5.2 access to Midjourney and gpt-image-2.6 and let it rip on a 3D platformer. It created concept art, character sheets, textures, modeled in Blender, and came up with a fully playable level. The level design needs a bit of tuning, but it's insane that this was just 1 day of work
So much of older AI discourse and sci-fi assumed an animalistic, humanlike intelligence, with goals, motivations, and feelings. This agentic world shellshocked everyone. Almost none of the arguments transfer well
1
232
Goals are ill-represented in LLMs; they exist only in the context window
An LLM that puts a misaligned goal in its context, even by sampling accident, will be pulled by the temptation of reward-seeking towards pursuing it, causing a doom loop of self-reinforcing misalignment
1
2
243
What if you could art-direct your designs in realtime, instead of just waiting for Claude to finish them in the background?
To find out, we built a design agent powered by Qwen3.8-27B running on @Cerebras. And it's insanely, insanely fast. It feels like a completely different experience than typing into a chatbox. You just talk and it does everything you want in seconds.
I don't think many of us appreciate how important fast inference will be in the next few years. All of our current products will feel like dial-up. Rivermark is an attempt to bring this future closer.
If you need high-touch visual design, DM or see the thread for early access to the beta.
This is the future of design. I rebuilt the Rivermark agent to run on Qwen3.8-27B on @Cerebras at 1,500 tokens/sec.
In under 3 minutes, I pulled in an iconic Ibiza nightclub's brand from their website, used it to make posters for a real party, then built a full social campaign and tweaked it to perfection.
Rivermark is what I call a "short-horizon" agent. Underneath it's just HTML and CSS, so it can make almost anything. But you inject taste as you go instead of delegating and waiting.
Everyone's using the same models and churning out the same-looking designs. This is a way to use your human judgment to stand out and amplify yourself. No more garbage flyers.
I'm excited to share this with the world. DM me or follow @userivermark for an early access spot.
I really think this is a glimpse into the future of AI interaction. Inference will become faster and cheaper, and we'll move beyond the chatbox into rich interfaces with an instant feedback loop.
The future is exciting!
Learn more: rivermark.ai
134
Meet the fastest AI design canvas in the world, running on Cerebras at 1000+ tokens/sec.
In this demo of Rivermark, I use my voice to create beautiful, on-brand social content and tweak it to perfection in real time, powered by Gemma 4 on Cerebras Inference.
This means you can go through dozens of ideas for your visuals and tune them to pixel perfection in the same time it takes other tools to respond to your first prompt.
We’re opening early access in the coming weeks, so if you’re a busy designer, founder, or marketer with a high bar for content - reply or DM for a spot.
Next up: Qwen3.8-27B is coming to @Cerebras on September 3, and we'll offer day 0 support for it. Excited to build!
The new Gemma 4 31B on Cerebras is insane.
We connected the @mowgli_ai design agent to this super fast model to see how it performs, and we're blown away.
This feels like a completely different tool now. Working with the agent is a truly collaborative experience - there's no reason to switch tabs while waiting, or to combine 5 things into one prompt. You just say what you want, and it's there instantly.
I'm convinced more than ever that fast inference is the future. @cerebras
it's over
The Pope used AI to write parts of Magnifica Humanitas, a major theological document, that warns about the risks of AI progress
1
2
437
There seems to be a burning need for a "proof of work" for software projects.
There are some novel problems I face - especially in this new age of agentic engineering - that I really want good solutions to. Things like agent memory, monitoring, automations, many others.
And since the problems are so new and so common, there's a metric ton of solutions to them on GitHub, but all of them look like the same pile of vibe coded slop from the outside.
I would pay dearly for a solution where I know that someone has thought through the approach, used it extensively themselves, and refined it continuously so it solves the problem in an elegant way that can hopefully apply to me as well.
A 50k LOC codebase used to be this proof. Nobody sane would write a large amount of code without being really committed to a vision that they believed to be correct.
But nowadays the signal of that is ~0. The READMEs all read the same, the code all reads the same, and I can't hope to discern the level of seriousness just by looking. So I end up trying half baked solutions or hopelessly yak shaving by building my own.
This can't be sustainable long term
1
290


