AI Engineering Reading Path
A practical AI engineering reading path for engineers, architects, and CTOs building AI systems in production. Explore how organizations structure context, connect knowledge sources, integrate tools, and govern AI agents at scale.
5 articles 49 min in total
Start with shared context
- 1
An Architecture for Vendor-Neutral AI Context Configuration
AI context configuration with AGENTS.md. Learn a vendor-neutral, monorepo-ready architecture that reduces lock-in compared to CLAUDE.md.
Build and package Agent Skills
- 2
Writing Agent Skills: Lessons Learned and Best Practices from Production Use
Learn how to design Agent Skills. Best practices from production covering maintainability, context efficiency, determinism, and reliability.
- 3
Packaging Agent Skills for Heterogeneous AI Coding Tools
Package Agent Skills and MCP servers once — then use them across ChatGPT, Codex, GitHub Copilot, Cursor and more with Agent Plugins 1.0.
Connect agents to data and documentation
- 4
Why AI Agents Should Use Database MCP Servers
Discover how database MCP servers reduce schema discovery effort, protect credentials, and improve observability for enterprise databases.
- 5
A Vectorless RAG Architecture for Online Documentation: Using Retrieval-Optimized Markdown Build Artifacts
Vector-database RAG struggles with raw online documentation. Explore an llms.txt-inspired approach using LLM-optimized Markdown artifacts.