AI Engineering Cheatsheets provides decision-ready references for common AI engineering problems, targeting practitioners who need structured guidance. It aims to streamline the selection of AI techniques, models, architectures, and production configurations, thereby accelerating problem-solving and enhancing the quality of AI-generated outputs.

How It Works

The project offers a collection of curated cheatsheets, each addressing a distinct area of AI engineering. Users are guided to open the cheatsheet relevant to their specific challenge, locate their situation within provided decision tables, and then follow the recommended approach. This methodology supports decisions on AI techniques, agent architectures (workflow, single, multi-agent), and optimizing LLM output via detailed prompt templates and review workflows.

Quick Start & Requirements

No explicit installation commands are provided; users are expected to access and utilize the cheatsheet documents directly. The primary requirement is access to Large Language Models (LLMs) and a foundational understanding of AI engineering principles. The repository links to external Towards AI courses and webinars for users seeking more in-depth knowledge.

Highlighted Details

• AI Engineering Playbook: Features decision tables for selecting appropriate AI techniques, models, prompting strategies, RAG setups, memory patterns, evaluation methods, and production configurations.
• Agent Architecture Guide: Offers a structured framework to decide between workflow, single-agent, and multi-agent systems, including essential questions and engineering rules.
• Anti-Slop AI Writing Guide: Includes a comprehensive 7-section prompt template, a list of banned words, style guidelines, and a two-model write-then-review workflow designed to produce human-like LLM output.

Maintenance & Community

This repository is maintained by Louis-Francois Bouchard, with links provided to his X, YouTube, and other AI resources. Additional learning materials are available through Towards AI courses and webinars.

Licensing & Compatibility

The provided README content does not specify a software license or offer compatibility notes for commercial use or integration with closed-source projects.

Limitations & Caveats

The cheatsheets function as reference guides, requiring users to interpret their specific scenarios within the provided decision tables. The effectiveness of the "Anti-Slop" guide is contingent on the LLM's adherence to complex prompt instructions. No explicit mention of alpha status, known bugs, or unsupported platforms is present.