How Intercom 2x’d their engineering velocity in 9 months with Claude Code | Brian Scanlan

2026-04-13 · Source: Lenny's Newsletter · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems · Depth: Advanced, extended

Summary

Intercom, a 15-year-old SaaS company, has doubled its engineering throughput, measured by pull requests per R&D head, in nine months by deeply integrating AI, specifically Claude Code, into its development workflow. This achievement stems from a company-wide focus on AI adoption, treating the engineering organization as a product, and leveraging deep telemetry to track AI tool usage and identify bottlenecks. Intercom's approach includes developing internal AI skills, enforcing quality standards for pull request descriptions using LLM judges, and actively addressing technical debt. The company views AI as a transformative force, enabling engineers to tackle complex problems faster and fostering a culture of rapid iteration and innovation, despite the increased operational costs associated with extensive token consumption.

Key takeaway

For Directors of AI/ML and VPs of Engineering aiming to significantly boost R&D velocity, prioritize deep AI integration across your development lifecycle. Instrument your internal AI tools with comprehensive telemetry to gain actionable insights into usage and bottlenecks, and actively cultivate a culture that empowers engineers to experiment and build AI-driven solutions. This approach can accelerate feature delivery, improve code quality, and tackle long-standing technical debt, fundamentally changing your team's output and morale.

Key insights

Deep AI integration and telemetry can double engineering throughput and improve code quality.

Principles

• Treat your engineering organization like a product.
• AI magnifies both strengths and weaknesses.
• Empower engineers to experiment and take risks with AI.

Method

Instrument internal AI skills with telemetry (e.g., Honeycomb), collect and analyze anonymized session data (e.g., S3, Snowflake) for insights, and build a repository of high-quality, evaluated internal skills with enforced standards.

In practice

• Use LLM judges to enforce PR description quality.
• Develop skills to automate tech debt reduction.
• Design CLIs and APIs to be agent-friendly.

Topics

• Engineering Velocity
• AI-Powered Development
• Claude Code
• Telemetry & Analytics
• Internal AI Skills

Best for: Software Engineer, Director of AI/ML, VP of Engineering/Data

Related on AIssential

• How Intercom 2X'd engineering velocity with Claude Code | Brian Scanlan — Intercom doubled R&D throughput and improved code quality by fully integrating AI-assisted development and treating internal AI adoption as a product.
• 🎙️ This week on How I AI: How Intercom 2x’d their engineering velocity with Claude Code — Intercom doubled engineering velocity in nine months by deeply integrating Claude Code, emphasizing telemetry and cultural permission.
• Beyond Linting: How AI Code Review Agents Are Learning to Think Like Senior Engineers — AI code review agents are evolving to provide rapid, senior-level architectural and security insights, not just syntax checks.
• 🧠 Community Wisdom: What to do when non-PMs start shipping directly to production, thoughts on Claude Code’s pricing A/B test, the use of gen AI in games, and more — Community discussions offer practical strategies for LLM optimization and production deployment.
• v2.1.80 — Recent Claude Code updates enhance agent capabilities, user experience, and system stability through new features and critical bug fixes.
• Meet the winners of our Built with Opus 4.6 Claude Code hackathon — Non-developers can create impactful AI applications using advanced coding assistants like Claude Code.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Lenny's Newsletter.