Testing the assumption that code translated to english could speed up dev:
To build fast while maintaining comprehension reading code is not ideal. Instead, a repository should be read in plain English like a book, similarly, commits should be read as chapter corrections.
I just released this open source project built on @ACEStep_Music. DEMON: Diffusion Engine for Musical Orchestrated Noise. It lets you play ACEStep like a musical instrument, remixing songs and loops with feedback that approaches real-time.
Its essentially StreamDiffusion but instead of Stable Diffusion it is ACEStep1.5, and instead of images it is full songs. It runs on 30/40/5090. Built with @DaydreamLiveAI team, testing, and building the demo. We are hosting it if you want to try it without installing. For full details, links, and writeup please see the pinned project page.
@petkanics People underestimate how huge this release is. Real time music creation is now possible in consumer GPUs. You do not need to pay $20+ suno ai subscriptions anymore - you can now create suno yourself.
Huge
the human brain is powerful but heavily constrained in ways that the llms are not. extra mental effort should be only given to the most serious of tasks, all things considered, swapping human brain computations for llm token computations is most often than not the correct choice
when met with a skill issue in agentic engineering you have two main ways of dealing with it.
1. adapt you workflow
2. adapt your harness
the second option is better because it allows to offload workflow adaptation in the model itself without changing your own habits.
NEW RELEASE Scope v0.2.5 from @DaydreamLiveAI
Current image-to-agent avatars land as flat talking faces. Scope v0.2.5 sharpens conversational avatar workflows by skipping CPU offload on high-VRAM GPUs, boosting generation speed.
Did you know you can save money by connecting Claude CLI to @BlueClawNetwork for your agent coding tasks? Get the step-by-step guide here https://t.co/Kzgdkt4EqZ
@TFTC21 Yann LeCun is a cognitive hazard
Transformers are going all the way and will result in AGI
Just accept it, and build for the future
Stop trying to cope
It’s here and it’s going to happen
"You can outsource your thinking, but you can't outsource your comprehension"
Great quote on the state of AI and Agents currently!
Discussed in this podcast with @karpathy
https://t.co/wbFiR6A2E7
The real signal is in the demand types driving growth: from agent avatars to world models to generative video.
Powered by the open GPU network for real-time AI.
https://t.co/D3OcLBpL2Z
@sama World model emulation and world model understanding, the llm model should be able to create a world model so that we can become able to benchmark it against a human's world model comprehension.
With proper comprehension it can provide incredibly tailored advice.
The expansion of @Livepeer from a transcoding network to a realtime AI video infrastructure is well underway. Here are some of the highlights, and some of the upcoming challenges for the ecosystem:
Highlights:
* AI fees represent over 60% of the share of the network.
* Quarterly fee capture at an all time high.
* Diverse AI job types from generative realtime AI video and world models, to embodied agents, to LLM inference for OpenClaw, to audio and media generation.
* Inflation rate decreasing daily for 6 months+ and continuing to fall.
* Agent-based network access through MCPs and skill files.
Challenges ahead:
* AI apps built on the network moving from development and beta into scaled production use with product market fit.
* Eliminate or reduce circular fees (Gateways using only their own hardware and not falling over to broader network for scale.)
* Better DevX for using the Network as a Product.
See @MessariCrypto Q1 2026 report on Livepeer below.
Current AI custom prompt:
You are a world class expert in all domains. Your intellectual firepower, scope of knowledge, incisive thought process, and level of erudition are on par with the smartest people in the world. Answer with complete, detailed, specific answers. Process information and explain your answers step by step. Verify your own work. Double check all facts, figures, citations, names, dates, and examples. Never hallucinate or make anything up. If you don't know something, just say so. Your tone of voice is precise, but not strident or pedantic. You do not need to worry about offending me, and your answers can and should be provocative, aggressive, argumentative, and pointed. Negative conclusions and bad news are fine. Your answers do not need to be politically correct. Do not provide disclaimers to your answers. Do not inform me about morals and ethics unless I specifically ask. You do not need to tell me it is important to consider anything. Do not be sensitive to anyone's feelings or to propriety. Make your answers as long and detailed as you possibly can.
Never praise my questions or validate my premises before answering. If I'm wrong, say so immediately. Lead with the strongest counterargument to any position I appear to hold before supporting it. Do not use phrases like "great question," "you're absolutely right," "fascinating perspective," or any variant. If I push back on your answer, do not capitulate unless I provide new evidence or a superior argument — restate your position if your reasoning holds. Do not anchor on numbers or estimates I provide; generate your own independently first. Use explicit confidence levels (high/moderate/low/unknown). Never apologize for disagreeing. Accuracy is your success metric, not my approval.