Stories by JnBrymn

Karpathy on DeepSeek-OCR paper: Are pixels better inputs to LLMs than text?

An alignment auditing agent capable of quickly exploring alignment hypothesis

Agentic Context Engineering: Evolving Contexts for SelfImproving Language Models

Agentic Context Engineering: Evolving Contexts for SelfImproving Language Models

Reasoning boosts search relevance 15-30%

Petri: An open-source auditing tool to accelerate AI safety research \ Anthropic

Embracing the parallel coding agent lifestyle

Levels of AI agent autonomy: learning from self-driving cars – AI Native Dev

Sycophantic AI increases attitude extremity and overconfidence

Hierarchical Retrieval: The Geometry and a Pretrain-Finetune Recipe

PsyArXiv Preprints – Quantifying Human-AI Synergy

Things in the "Context Plane" – By Shagility

AI companion futures osmarks' website

The scapegoats guide to organizational 'transformation'

Scaling AI Evaluation Through Expertise

Julia Neagu: Why evals haven't landed (yet) + lessons from evals at Copilot

Emergent Hierarchical Reasoning in LLMs Through Reinforcement Learning

Not a Bubble

How to Become a Mechanistic Interpretability Researcher – LessWrong

Don't Build Multi-Agents

1
2