# Tammuz Dubnov - When Our PM Started Writing Code: What Merge Rate Taught Us About AI Adoption - AI N

## Executive summary

The shift toward 'AI native' organizations means that the primary bottleneck is no longer code generation speed but organizational alignment and handoff efficiency. The speaker argues that empowering non-technical decision-makers (PMs, Designers) with agentic tools allows them to execute work directly, collapsing gaps that previously required multiple sprints or lengthy coordination. Success in this transition is measured by the 'merge rate'—the percentage of Pull Requests (PRs) opened by non-technical users that successfully land in production.

## Key takeaways

- Redefining AI Native: AI native means that the person who cares about a feature has the authority and ability to execute the work, collapsing the traditional handover gap between PMs/Designers and Developers. The bottleneck shifts from coding speed to decision-making capacity.
- The Importance of Merge Rate: Merge rate (the percentage of PRs that land in production) is the key metric for measuring if an organization is successfully adapting to AI-driven workflows. A high merge rate indicates trust and quality between non-technical contributions and the dev team.
- The Role of Guards and Architecture: As organizations move fast, it is critical not to abandon engineering principles or 'guards' (like testing, architectural standards). The agent must be designed to learn the codebase deeply and maintain these constraints.
- Measuring Non-Technical Contributions: To measure impact, track: 1) Count of PRs opened by non-technical individuals. 2) Merge rate (e.g., an average merge rate of 74% was cited). 3) Percentage of merged PRs that require zero developer intervention.

## Technical details

- AI Native Definition: The core principle is enabling the person who makes the decision to also execute the work, bypassing traditional handoffs (PM -> Figma -> Dev sprint). This process can collapse weeks of waiting into minutes.
- Agentic Workflow Principles: Effective agents must be coupled deeply to the codebase, understand product-level language, manage long sessions, and constantly check their own work for automatic feedback loops. The system should adapt to prevent repeated mistakes.
- Measuring Developer Burden: To minimize PR fatigue on the dev team, agents should be optimized to output minimal lines of code by maximizing reuse and modeling dependencies. Furthermore, internal tools can label incoming PRs with a risk level and size for better load management.
- Development Practices: The speaker recommends using 'feature flags' extensively to allow developers to make product decisions quickly, enabling multiple follow-up PRs (e.g., three or four) for cleanup without accumulating conflicts.

## Practical implications

- Shift focus from increasing LLM compute spending to empowering decision-makers (PMs, Designers) with specialized tools.
- Implement robust CI/CD practices and 'guards' to ensure that rapid AI development does not compromise core engineering principles or product stability.
- Adopt metrics beyond just PR count; measure the merge rate and the percentage of merged PRs requiring zero developer commits to validate process efficiency.
- Utilize specialized tools for non-technical users rather than giving them general 'cloud code' access, which can lead to nonsense PRs.

## Topics

AI Adoption, Product Management (PM), Merge Rate Metrics, Agentic Workflows, Build Process Optimization, The AI Native Dev Podcast

Source: https://www.youtube.com/watch?v=KzhjnILSP0Y