Refactoring Podcast

Cate Huston, engineering leader and coach who’s led teams at DuckDuckGo and Automattic. Jean Hsu, fractional engineering leader and coach who builds career programs for startups. They talk about owning your career intentionally, making coaching scalable and affordable, using journaling and cohort structures to sustain growth, mapping personal goals to workplace realities, and how AI changes tools but not core career work.

Stuart Caborn, distinguished engineer at loveholidays who builds platform engineering and AI-first developer workflows. He discusses applying AI to real-world software delivery, keeping elite code health while 60%+ of production code is AI-assisted. Short, concrete stories cover data products, guardrails, skills-as-executable-docs, and how they measure and scale safe AI-driven change.

Chris Lattner, creator of LLVM, MLIR, and Swift and founder of Modular, rethinks AI infrastructure for portable, heterogeneous compute. He explores making GPUs accessible, the new compiler stack (Mojo/MLIR), and how AI shifts engineering workflows. Conversations cover open source pressures, junior engineers thriving with AI, and the need to think long-term about maintainability and tooling.

Amelia Wattenberger, product lead at Augment Code and former principal research engineer at GitHub Next, discusses post-IDE workspaces and AI-driven developer tooling. She explains Intent’s model of isolated workspaces, agent personas and orchestration. Conversation covers specs as a control plane, opinionated interfaces beyond chat, notes for collaboration, and how AI reshapes engineering roles.

Richard Hughes-Jones, a seasoned executive coach for founders and senior tech leaders, talks leadership, uncertainty, and AI. He explains coaching without an engineering background. He explores leading beyond frameworks, holding polarities, and designing safe-to-fail experiments. He describes using AI to augment coaching with transcripts, pattern spotting, and reminders.

Austen McDonald, former senior engineering manager and hiring committee chair at Meta and author of Mastering Behavioral Interviews, shares why behavioral interviews matter now. He breaks down the decode-select-deliver loop, the eight signal areas interviewers seek, the big three prompts, building a story catalog, and ethical framing. Short, practical, and aimed at both candidates and recruiters.

Paige Bailey, Developer Relations Lead at Google DeepMind and ex-Principal PM at GitHub who helped launch Copilot, demos building a webcam app entirely with voice. She walks through removing backgrounds, running a live AI interview to auto-fill profiles, choosing models, deploying to Cloud Run, and when to move from Studio to IDEs. The conversation covers team workflows, agent-driven automation, and how roles are evolving with AI.

Gonçalo Silva, the CTO of Doist, leads a unique company with 100 remote employees across 40 countries, creating the beloved productivity tool Todoist. He shares insights on staying small and effective, contrasting bootstrapped constraints with the pressures of VC funding. The discussion dives into the origins of Doist's async-first culture, how they manage team dynamics, and balance strategy with user feedback. Gonçalo also explores the role of AI in enhancing productivity, emphasizing clarity and innovation within their teams.

Thiago Ghisi, a former engineering director at Nubank and Apple, shares his transformative sabbatical journey. He reveals how detaching from his work identity allowed him to embrace self-discovery and deeper questioning. Thiago discusses the creation of a 'second brain' using tools like Obsidian and Cloud Code to analyze 15+ years of notes for insights into his past. He highlights the importance of sustainable routines and the unique challenges of finding joy outside productivity. Finally, he considers pursuing a PhD in psychology, blending his tech skills with newfound passions.

In a dynamic conversation, Matt McClernan, CEO of Augment Code and an AI coding expert, dives into the state of AI adoption among engineering teams. He explores the gap between personal and team adoption, stressing the role of documentation as a powerful tool for enhancing AI outputs. Matt reveals how context engineering, particularly through RAG-based retrieval, can transform coding practices. He also discusses the future of AI in developer workflows, including the evolution of UX tools like IDEs and CLIs and the importance of governance in this rapidly changing landscape.