Encoding Team Standards

2026-03-31 · Source: Martin Fowler · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning · Depth: Advanced, long

Summary

AI coding assistants often produce inconsistent output quality because their responses depend on individual developers' ability to articulate team standards, often based on tacit knowledge. The article, published on 31 March 2026, proposes treating AI interaction instructions (for generation, refactoring, security, and review) as shared infrastructure. This "Executable Governance" approach involves creating versioned, reviewed artifacts that encode a team's implicit judgment into explicit, executable instructions. This ensures consistent application of team standards, regardless of who is prompting the AI. A well-structured instruction includes a role definition, context requirements, categorized standards, and a defined output format. This system applies across the entire development workflow, from initial code generation to continuous integration, and is particularly beneficial for teams of fifteen or more to scale quality and reduce bottlenecks.

Key takeaway

For AI Architects or MLOps Engineers aiming to standardize AI-assisted development, you should implement versioned, executable AI instructions. This approach transforms senior engineers' tacit judgment into consistent, team-wide standards for code generation, refactoring, and security checks. By embedding these instructions into your workflow, you can reduce inconsistency, scale quality across your team, and prevent junior developers from becoming bottlenecks. Consider starting with one high-value instruction, like generation or review, to demonstrate immediate impact.

Key insights

Encode team's tacit judgment into versioned, executable AI instructions to ensure consistent quality across developers.

Principles

• Treat AI instructions as shared infrastructure.
• Shift tacit team knowledge to explicit, executable rules.
• Embed governance directly into the development workflow.

Method

The process involves interviewing senior engineers to extract tacit knowledge, then structuring it into AI instructions with role, context, categorized standards, and output format.

In practice

• Implement generation instructions for new code.
• Use security instructions for threat model checks.
• Apply review instructions for consistent quality gates.

Topics

• AI Coding Assistants
• Team Standards
• Executable Governance
• Tacit Knowledge
• Software Development Workflow
• Code Quality

Best for: CTO, VP of Engineering/Data, AI Engineer, AI Architect, Director of AI/ML, MLOps Engineer

Related on AIssential

• AI tools for hearing difficulties — helpful or harmful for language learning? — AI assistive technology presents a dilemma for language learners with hearing difficulties, balancing aid with skill development.
• How Enterprise Marketing Teams Are Actually Using AI Agents in 2026 — Enterprise marketing is shifting from human-assisted AI tools to autonomous AI agents that redesign entire workflows.
• [AINews] Microsoft Build: MAI-Thinking-1 and MAI Family models — Microsoft's MAI models showcase a transparent, full-stack AI strategy emphasizing clean data and local agent execution.
• Are we getting better at explaining things because of our interaction with LLMs? — Crafting clear LLM prompts may enhance human explanatory and communication abilities.
• OpenAI expands Codex with role-specific plugins to build a general-purpose app for non-developers — OpenAI is transforming Codex into a general-purpose, role-specific AI work app, expanding beyond coding for non-developers.
• How to Use Claude Managed Agents? — Anthropic's Claude Managed Agents simplifies AI agent deployment by providing a hosted, stateful, and secure execution environment.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Martin Fowler.