We're sharing our internal AI engineering cheatsheets

2026-01-08 · Source: Learn AI Together · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, quick

Summary

Towards AI has released a set of three markdown files containing distilled engineering decisions, agent architecture, and writing quality guidelines, based on years of experience and tested on real systems. These resources are designed to provide decision-ready references for common AI engineering problems, such as model selection, agent structuring, and generating natural-sounding output. The content is derived from Towards AI Academy courses and aims to enable rapid development, similar to Anthropic's recent rapid shipping of features like Cowork, direct computer use, persistent threads, plugins, and scheduled tasks. The files are available on GitHub and can be integrated directly into AI development workflows, specifically by providing them to large language models like Claude Code for contextual guidance and automated review.

Key takeaway

For AI Engineers building with large language models like Claude Code, integrating these decision-ready markdown files can significantly streamline your development process. By providing Claude with pre-defined engineering and agent architecture guidelines, you can ensure more consistent and higher-quality outputs, reducing iteration time and improving the naturalness of generated text. You should add the engineering and agent files to your LLM's context and the writing guide as a skill step for automated review.

Key insights

Pre-made engineering decisions and architectural guidelines accelerate AI development and improve output quality.

Principles

• Accumulated judgment drives rapid AI development.
• Contextual guidance improves AI model performance.

Method

Integrate markdown-based engineering and agent architecture files into an LLM's context, and use a writing guide as a skill step for automated output review.

In practice

• Use provided cheatsheets for model and setup decisions.
• Structure agents using the recommended architecture.
• Review AI generations against the writing guide.

Topics

• AI Engineering
• Agent Architecture
• Claude Code
• Engineering Best Practices
• AI Development Cheatsheets

Code references

• louisfb01/ai-engineering-cheatsheets

Best for: AI Engineer, Machine Learning Engineer, Prompt Engineer

Related on AIssential

• The engineering best practices you can drop straight into Claude — Towards AI offers free, public AI engineering cheatsheets to accelerate LLM system development.
• Claude Code Design Patterns for AI Agents — Claude Code's architectural primitives enable engineering reliable, production-grade AI agent systems by applying classic software design patterns.
• 🎙️ This week on How I AI: How to build your own AI developer tools with Claude Code — Custom AI-native dev tools and model-to-model evaluation enhance AI engineering workflows and code quality.
• v2.1.91 — Recent Claude Code updates enhance security, performance, and developer experience through new features and extensive bug fixes.
• Level Up Your Claude Code with This CLAUDE.md — A "claude.md" file defines an AI assistant's engineering workflow, ensuring consistent, disciplined code generation.
• Using Claude Code: The unreasonable effectiveness of HTML — HTML offers richer, more readable, and interactive outputs for AI agents compared to Markdown.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Learn AI Together.