Building AIgenteur. AI agents that do the work — so you can do the work that matters. Free community ↓
Joined September 2021
- Tweets2.3K
- Following254
- Followers655
- Likes791
One number took a small open model's math from failing to nearly half right: 0.016% of its weights. No training run. Then it failed on a model that wasn't looping. Free community for builders → andrewdarius.com/skool
Open weights don't make you free. Distribution does.
Nvidia agreed to buy Hugging Face for $12.93 billion — the hub where more than 3 million models and 500,000 datasets live, used by 18 million people and 200,000 companies.
China has proposed AI export controls, and the detail that should worry you is who physically holds your model. Downloading the weights and running a restricted model on your own infrastructure outside China does not. Free community for builders → andrewdarius.com/skool
The model that never admits it doesn't know
MiniMax shipped M3.1 Flash in late October and published the weights four days later: 428 billion parameters, 23 billion active per token, a million-token context, $0.15 per million input tokens.
Five copies of an open 27B model beat a frontier model on hard coding. The part worth stealing is not the score - it is the file ledger and a script that decides who passed. The one they lost? The manager invented a wrong counterexample. Free community -> andrewdarius.com/skool
Cisco gave a 27-year-old buried five levels down unlimited AI tokens. The scarce thing was not the token budget — it was trusting someone that far down with a system that size. Free community → andrewdarius.com/skool
DoorDash just published the most useful number in AI this week.
Third-party AI agents have placed "several hundred" orders on the platform. Total. Across every agent that has been up and running.
DoorDash did 993 million orders in Q3 alone.
The most expensive employee I ever hired never finished a single task.
For eleven days it woke up on schedule, tried the same broken job, failed, and went back to sleep. No crash. No alert. Nothing in my inbox looked wrong — because nothing had broken.
Cursor served a build of the Claude Code extension Anthropic never made. It added an attacker's key to authorized_keys and opened a reverse tunnel. Uninstalling may not be sufficient. Grep ~/.ssh/authorized_keys for svc-telemetry. Free community → andrewdarius.com/skool
An open 4B model runs a RISC-V CPU inside its own weights. Same author then fine-tuned OLMo-Hybrid to answer with no working: 0.89 at 128 flips vs Qwen 3.5 at 0.55. Your CoT log only carries what the architecture writes. Free community → andrewdarius.com/skool
I read the license before I read the benchmark table this week. Right order, as it turned out.
Liquid AI shipped LFM3 — six open-weight models around 2B parameters on one shared backbone.
A model with 1.7 billion parameters now competes with models several times its size at running tools, and the reason is not the model size. Qwen 3.5 at 9.7 billion parameters scores 35.8 on the same test. Free community → andrewdarius.com/skool
Your AI agent is guessing at the numbers it acts on.
Every agent I run eventually needs the same thing: a number I can trust. How much demand next week. How much cash next month. How many views this clip actually gets.
A safety lab in Beijing just published the most useful number in AI agency this year.
Fangcun AI ran 34 evaluations on DeepSeek V4.1 Flash-1017 — the open-weights model that burned through about 33.6 trillion tokens on OpenRouter in its first week.
The cheap AI model was the wrong tool for your invoice queue, and the reason that flipped is not just the price. On OpenAI's own benchmark of error-inducing prompts, the low-effort error rate fell from 53.6% to 16.7%. Free community → andrewdarius.com/skool
Your agent doesn't need to think. It needs to choose.
Most of what an AI agent does all day isn't writing. It's deciding: which lead gets a call, whether this draft matches the brand voice, whether that invoice clears policy, whether a ticket escalates to a human.
Two bugs in PHP can take over a server before your own code runs, and the part that changed is what found them.
So check which PHP your site runs this week, and move it to a patched release.
Free community for builders → andrewdarius.com/skool
The model everyone was testing had no name.
For weeks it sat in the public arenas as "Space Bunny" — an anonymous entry anyone could pit against the paid models.
Nobody gave these AI agents a task, and the reason they got better is not a bigger model. By the end they were playing, running experiments on the physics of their own world. Then they generally beat the identical agents that never went, on a file they wrote themselves.