Stories by vinhnx

JPEG Compression

Autoresearch Home

From model to agent: Equipping the Responses API with a computer environment

The Artificial Self

Taming chaos is a learnable skill

Ways Product Discovery Breaks Down

Reinforcement Learning environments and how to build them

Slate: Moving Beyond ReAct and RLM

The Marginal Hire

Xterm Control Sequences (1994)

I made a real BMO local AI agent with a Raspberry Pi and Ollama [video]

I upgraded my BMO local AI agent's voice and brain [video]

The Film That Should Be Nominated

Apple at 50: Five Decades of Thinking Different [video]

Nvidia Nemotron 3 Super

10x Is the New Floor

I believe in SOTA models over custom ones

Ripgrep Code Review (2016)

10x Is the New Floor

Teaching LLMs to reason like Bayesians

1 2 3 4 5 6 7 8 9 10