Running an AI-native engineering org

2026-06-03 · Source: Claude Blog · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning, Operations & Process Management · Depth: Intermediate, medium

Summary

At Code w/ Claude SF 2026, Fiona Fung, Director of Engineering for Claude Code, detailed how her team transitioned to an "AI-native" engineering organization, making agentic coding the default. This shift moved bottlenecks from code writing to verification, code review, and security. The team rewrote norms, adopting just-in-time (JIT) planning over six-month roadmaps, and now asks Claude for code context instead of authors. Human code review focuses on critical areas like legal, security, and product sense, while Claude handles style and basic bug fixes. Team roles have blurred, with PMs coding and engineers taking on design, leading to a hiring focus on creative builders and deep systems experts, rather than raw throughput. Core principles include dogfooding Claude Code, maintaining a flat team structure, and actively eliminating obsolete processes. Engineering leaders should track onboarding ramp time, PR cycle time, and Claude-assisted commits to gauge the effectiveness of these new norms.

Key takeaway

For engineering leaders transitioning to AI-native development, you must proactively question and eliminate obsolete processes that hinder agentic coding's benefits. Shift your planning to just-in-time methods and re-evaluate human involvement in code review, focusing expertise on critical areas like legal and security. Prioritize hiring creative builders and deep systems experts, and track new metrics like onboarding ramp time, PR cycle time, and AI-assisted commits to measure your organizational agility and effectiveness.

Key insights

Agentic coding fundamentally redefines engineering bottlenecks, shifting focus from code generation to verification and strategic human oversight.

Principles

• Relentlessly dogfood your product.
• Maintain a flat team structure.
• Actively eliminate obsolete processes.

Method

Implement just-in-time planning with prototypes and internal feedback. Automate context gathering by querying AI first. Focus human code review on legal, security, and product expertise, while AI handles style and basic bug fixes.

In practice

• Automate customer feedback summaries.
• Monitor onboarding ramp time and PR cycle time.
• Question and potentially automate "noisy" workflows.

Topics

• AI-Native Engineering
• Agentic Coding
• Workflow Automation
• Just-in-Time Planning
• Code Review
• Engineering Metrics
• Team Reorganization

Best for: Director of AI/ML, MLOps Engineer, AI Engineer

Related on AIssential

• A Practical Guide to Becoming an AI-Native Engineer — AI-native engineering requires orchestrating AI agents with disciplined practices and robust security, shifting focus from coding to verification and context.
• A Year Of Vibes — Agentic coding tools are fundamentally reshaping software engineering workflows, challenging traditional practices and human-machine relationships.
• Personal leverage, junior engineers, and weekly reading 💡 — AI tools empower individual engineers to achieve unprecedented output, while junior engineers bring unique "AI-native" skills to development teams.
• Conductors to Orchestrators: The Future of Agentic Coding — Software engineering is shifting from direct coding to orchestrating autonomous AI agent fleets for scaled productivity.
• From Coder to Manager: Navigating the Shift to Agentic Engineering with Notion Co-Founder Simon Last — Notion is evolving into an agent-centric platform, shifting from human-direct work to human-managed AI agents.
• How I code with agents, without being 'technical' — AI agents enable non-technical users to build complex software by abstracting traditional coding into a guided, iterative process.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Claude Blog.