Human-AI Collaboration Model
In one line: The human owns what and why; the AI owns how — under human validation and review.
What: AI-assisted development is a collaboration between an architect (human) and a senior engineer (AI), not automation. The human decides what to build and why, and validates proposals against requirements, constraints, and architectural coherence. The AI proposes how, implements, and the human reviews.
Why: Two opposite failure modes motivate the split.
AI without judgment. Unsupervised, the AI produces code that is locally correct but globally incoherent — each component works alone, the system doesn't compose. It lacks business context and long-term vision, optimizes for the immediate request, and ignores cross-cutting concerns.
Human without velocity. A solo architect implementing a complex system by hand spends months after the design is done. The architecture is sound but the time-to-market is not. Judgment is irreplaceable; typing speed is a bottleneck.
The model resolves both: the human supplies judgment, context, and architectural vision; the AI supplies implementation velocity, comprehensive test generation, and consistent pattern adherence without quality decay.
Specific Division of Labor:
| Responsibility | Owner | Rationale |
|---|---|---|
| Problem definition | Human | Requires business context the AI does not have |
| Approach selection | Human (with AI proposals) | AI proposes 3+ approaches; human selects based on constraints AI cannot fully evaluate |
| API contract design | Human | Contracts affect other teams and external consumers; judgment-intensive |
| Implementation | AI (under human review) | Pattern-following work where AI excels |
| Test implementation | AI (under human review) | Comprehensive test generation is an AI strength; coverage gaps are a human blind spot |
| Code review | Both (two-stage) | AI reviews for spec compliance and patterns; human reviews for architectural coherence |
| Regulatory compliance | Human (with AI support) | Legal interpretation requires human accountability; AI assists with completeness checks |
| Documentation | AI (under human review) | Volume work where AI maintains quality; human ensures accuracy of claims |
| Refactoring | AI (under human direction) | Mechanical transformation where AI consistency prevents errors; human decides what to refactor |
| Architecture decisions | Human | ADRs require judgment about long-term tradeoffs |
Evidence: The division is observable in the commit log: the Co-Authored-By trailer marks AI-paired work, and the human-only commits cluster on config and manual interventions. Reproducible by inspecting trailers across the history.
How: The collaboration model is encoded in infrastructure, not left to discipline:
-
CLAUDE.md (project-level) provides the AI with architectural context, technical constraints, and established patterns. This is the primary mechanism by which the human architect's judgment persists across AI sessions. A well-written CLAUDE.md is worth more than any prompt engineering technique because it operates at the system level rather than the request level.
-
Memory files (
memory/MEMORY.mdand linked files) provide the AI with accumulated knowledge about the project's history, preferences, known debt, and regulatory requirements. This prevents each new session from starting at zero. -
Superpowers skills enforce the lifecycle structure. The AI cannot skip the design phase or the verification phase because the skills define the workflow sequence.
-
Agent-based review (Section 5) provides automated second opinions. The human architect is not the sole quality gate — specialized reviewer agents check for security issues, API consistency, and compliance gaps.
-
Subagent-driven development ensures fresh context per task. Rather than one long session that accumulates context pollution, each implementation task dispatches a fresh subagent with only the relevant context. The orchestrating session reviews output between tasks.
For the full agent architecture, see Section 5. For memory system details, see appendix-d-memory.md.