Bob Martin: Don't Build Big Architecture Upfront for AI Agents
mariofilhoml · x · 2026-07-30
Renowned software engineer Bob Martin (Uncle Bob) shared his architectural philosophy for working with AI agents. He stated that he is not a fan of big architecture upfront, preferring to let agents run for a while and adjust the architecture iteratively.
He typically provides only basic hints upfront, such as keeping dependencies pointing from low-level to high-level modules, allowing the architecture to evolve naturally. Furthermore, he believes that if a developer has to manually intervene and inspect the code, it should be viewed as a failure, prompting the need to figure out what went wrong and invent new mechanisms to prevent it in the future.
More from coding & agent
- Engineering Management: Banning Direct AI Outputs in PR Discussions — braelyn_ai · 2026-07-30
- CORTEX: Open-Sourcing Mechanistic Interpretability for Local LLMs — JayB_Official · 2026-07-30
- flujo: An Open-Source MCP Server Manager to Kill JSON Config Hell — Ambitious-Prompt-975 · 2026-07-30
- Beware AI Vendor Lock-in: Encrypted Reasoning Tokens Strip User Control — mitsuhiko · 2026-07-30
- Is Agentic Coding Actually Useful? Senior Engineers Weigh In — ParaboloidalCrest · 2026-07-30
- MoTA: Replace Massive Context with 4MB LoRA Adapters, Cutting Inference Storage by 100x — EyalToledano · 2026-07-30