In this episode, I sit down with Brittany Ellich, Staff Engineer at GitHub, to talk about what software engineering looks like in the age of AI. We explore her path from working on GitHub Actions to building metered billing systems for enterprise products, where accuracy matters more than speed and even small mistakes can have major business impact. Brittany shares how billing for usage-based products like Copilot and Actions becomes especially challenging in distributed systems, across cloud environments, enterprise configurations, and data residency boundaries.We also dig into how her day-to-day engineering workflow has changed with AI. Brittany explains how she splits work into synchronous and asynchronous modes, when she delegates tasks to cloud-based coding agents, and when she prefers working directly in VS Code with Copilot. We discuss agent-driven development, tools like Squad, the shift from writing code to shaping systems and products, and what all of this means for junior engineers, technical debt, productivity, and the future of software careers. It’s a practical and thoughtful conversation about learning to trust AI, staying adaptable, and figuring out what’s actually worth building now that creating software is faster than ever.In this episode, I sat down with Brittany Ellich, staff engineer at GitHub, to talk about what software engineering looks like when AI is involved in nearly every step of the workflow. Brittany shared her path from a smaller medical education company into GitHub, where she first worked on GitHub Actions and now focuses on enterprise billing and metered usage. We explored why billing is such a fascinating engineering problem: in her world, accuracy matters far more than speed, and even small mistakes in a distributed system can have major business impact. She also explained how GitHub handles the complexity of enterprise licensing, usage-based billing, and data residency across different environments.From there, we went deep on AI-assisted development. Brittany described how she splits her work between asynchronous tasks delegated to cloud-based coding agents and synchronous work inside her IDE with Copilot, and why the real challenge is no longer just building software, but deciding what is worth building. We also discussed tools like multi-agent "squad" workflows, the changing role of engineers as product and architecture skills become more important, the pressure this shift creates for junior developers, and why she remains optimistic about the future of software engineering. Brittany closed by sharing her growing interest in using AI for refactoring and technical debt reduction, and why that may be one of the biggest opportunities ahead.In this episode, I sat down with Brittany Ellich, Staff Engineer at GitHub, to talk about how AI is reshaping modern software engineering. We explored her path from traditional software roles into developer tools at GitHub, where she now works on enterprise billing and metered usage systems—an area where accuracy matters far more than speed. Brittany shared what it’s like to build reliable billing infrastructure at massive scale, why underbilling can be just as painful as overbilling, and how complex things get when you add distributed systems, SaaS products, enterprise licensing, and data residency requirements into the mix.We also dug into how Brittany actually works with AI day to day: delegating smaller, well-scoped tasks asynchronously to cloud agents, using Copilot and agent mode for more exploratory work, and experimenting with multi-agent tools like Squad to prototype ideas quickly. Our conversation covered the shift from “writing code” to “shaping systems,” why engineers are becoming more like product thinkers and architects, what this means for junior developers entering the field, and why Brittany remains optimistic about the future of software engineering despite all the uncertainty. If you’re wondering how AI is changing the real workflow of experienced engineers right now, this one is packed with practical insight.