AI tools, guidance, and research for AI engineering at scale.
Built from real-world experience optimizing AI workflows, this repository contains research-backed tools for context optimization, prompt engineering, and LLM-assisted development.
- Skills - Skills developed from research-backed information
- Prompt Patterns - Production-tested system prompts and templates
- Engineering Tools - Scripts and utilities for AI-assisted workflows
Most AI tooling is built on intuition. This repo is built on research - Google, Stanford, UC Berkeley, Anthropic, Microsoft, Open AI Research.
Stop cargo-culting. Start measuring fidelity, cache efficiency, and task success rates.
Built by Arthur Khachatryan