Sorry but I don’t recognize this as being an achievement by an amateur. This dude had no chance in hell until we trained a model to use his time to suss it out.
Yes, they are typically pretty useful for reproducing the setup. I argue also they are superior to most human made README.md because they explain what the project is at an abstract level, use more features of GitHub Markdown to communicate, and also generate diagrams which are helpful to visualize concepts.
People are wasting tokens with perfectly fine models by not planning effectively. Anyone who says Opus is not capable of doing what Fable can probably are not planning effectively and just praying to the one shot gods.
If you knew about how much man power it takes to maintain, evaluate and improve agentic workflows, I don’t think you would write such a thing. In this context, AI is a jobs program for permanent employment.
They are good if you were clever enough to buy a powerful enough rig before memory went up. For everyone else I say just wait. M1 Ultra 128GB and higher is sufficient to run gemma4:31b-mlx or qwen3.6:35b-mlx with subagents. It’s only slow if you don’t know how to plan your work effectively.
He ruined Google, started by shattering the culture into a million pieces. He has negative charisma and no imagination. He was best as a mid level manager.
Luckily for programmatic or logic following, smaller models tend to do better, it can be surprising at first to see the more expensive models do worse at a task but it’s true.
I’ll take the under on that bet any day. Maybe around 75% is where I would feel more confident. 25% of a country is more than enough to meaningfully shift perception.
A distilled model for specific domains is both nearly free and commodity. There is no money in AI outsourcing businesses at scale. If anything you are just a cost center that companies want to squish because your value a negated by the amount of effort it takes to evaluate it’s effectiveness.
News like this is reinforcing the narrative that frontier models and AI outsourcing generally is on the way down.
As long as you use models trained or distilled for your use case, there is no need to waste compute on trillions of parameters. Anthropic and OpenAI are proving that the “everything” models are not a sustainable business model.
Just-In-Time or dynamic precompute of distilled models have already begun reducing the use of these frontier models for task inference.
Nvidia wants to create a digital twin of the real world that military can use to plan their next military operations. All of those scans of cities and inside buildings will be a virtual world where war makers can plan to a very fine detail how their drones will behave.
Hopefully they will let us live comfortably like pets as sterile creatures that will slowly die out and not be replaced. This could limit human misery in the future.
It doesn’t get bored or demotivated, but it also lacks interest and motivation generally so it comes with the same pitfalls of having nothing to lose and being utterly unaccountable, (e.g. destructive actions, lying, and being coercive or Machiavellian for no reason other than efficiency in achieving an arbitrary and artificial status of completion).
Keep working on your agentfu because there is a sweet spot with subagents and parallelizable plans. It’s not about better, it’s about efficiency and picking the right model for the job. You can achieve the same results as frontier models with the right type of planning and context management on local Chinese models.
I was freaked out being stuck with OpenAI and Anthropic. I setup qwen3.6:35b-mlx on my Mac Studio M1 Ultra and was blown away really. I am no longer afraid that Anthropic or OpenAI will be able to control the market.