Pinned Loading

1. sabotage sabotage Public

   Heron AI Work Test | Project - Michael Chen Sabotage Threat Modeling. A small eval harness designed to measure Monitor Negligence. The script simulates a "Monitor" (the insider) reviewing dangerous…

   Python 1

2. Context-Engineering Context-Engineering Public

   "Context engineering is the delicate art and science of filling the context window with just the right information for the next step." — Andrej Karpathy. A frontier, first-principles handbook inspi…

   Python 8.1k 918

3. aisecforge aisecforge Public

   Open-source resources for systematic adversarial testing, evaluation, and security hardening of large language models. This repository consolidates novel methodologies for identifying, classifying,…

   6 1

4. quant-lab quant-lab Public

   An AI Quant Trading Lab

   Python 8 3

5. universal-deep-research universal-deep-research Public

   A minimal practical implementation of NVIDIA's Universal Deep Research system. https://arxiv.org/abs/2509.00244

   Python 4 1

6. RL101 RL101 Public

   Agentic Reinforcement Learning 101. A pragmatic course for AI/ML Engineers based on "The Landscape of Agentic Reinforcement Learning for LLMs: A Survey" https://arxiv.org/abs/2509.02547

   Roff 12 2