As software development scales from tens of thousands of human developers to millions of machine-speed developers, the assumptions behind traditional codebase management are being challenged. Practices that worked at one scale can behave very differently as development activity increases dramatically.

In this session, you will explore the consequences and trade-offs of managing Google’s multi-billion-line monorepo in this new environment. You will examine the advantages of a large shared codebase for global optimisation, alongside the scaling challenges created by years of organic growth and increasingly complex dependency relationships. The session covers build and test caching, dependency management, measuring the impact of codebase cleanup efforts, and techniques for reducing churn. You will also examine how codebase structure can help limit the blast radius of agentic development activity and the common misconceptions that can make this harder than it needs to be.

What You Will Learn

• The advantages and trade-offs of managing very large codebases as development scales to machine-speed workloads

• How build and test caching, dependency management, and cleanup efforts influence codebase scalability

• How codebase structure can reduce churn and limit the blast radius of agentic development activity

Who Should Attend

Software Architects, Platform Engineers, Developer Infrastructure Engineers, Build and Release Engineers, Staff and Principal Engineers, Technical Leads responsible for large codebases, and Engineering Managers overseeing software development platforms.