The Pragmatic Engineer
Stop being skeptical about AI for development with Charity Majors
· Summarized
Key takeaways
- AI is fundamentally altering software engineering, from code writing to verification and deployment. It offers tremendous efficiency but demands new discipline, especially regarding reliability, testing, and observability.
- The traditional artifacts of software engineering, like code reviews, need to evolve. Focus should shift towards system behaviors, API design, and system relationships, underpinned by robust observability and telemetry.
- Verification and trust in AI-generated code are crucial. Automation of testing and continuous monitoring will be key to managing non-deterministic systems.
- Leadership in tech must involve technical hands-on involvement, discipline, and empathy. Managers should embrace AI skills, return to coding when necessary, and foster a culture of rapid iteration and trust.
- Both camps of AI skeptics and enthusiasts are recognizing real trends—AI accelerates progress but also introduces challenges—effective communication about its costs and benefits is essential.
- Engineers and leaders should proactively develop AI fluency. Embracing experimentation and quick adaptation are vital to staying ahead in this rapidly changing landscape.
- The security, reliability, and ethical use of AI should be treated as core to modern engineering practice, with ongoing emphasis on safeguarding systems, managing risks, and maintaining human oversight.
Summary
The Evolution of Charity Majors' Perspective on AI in Software Engineering
Charity Majors describes her journey from skepticism to acceptance regarding AI's role in software development. Initially, in 2020, she thought of AI more as a feature than a foundational shift, but by 2025, particularly after the release of Opus 4.5, her viewpoint shifted as she recognized AI's potential to fundamentally change how code is generated and verified. She emphasizes that AI's strengths lie in writing code and its incredible ability to automate and optimize processes, but also highlights the importance of verification and discipline, especially given the non-deterministic nature of these systems.
Implications of AI for Software Development and Reliability
The podcast discusses how AI impacts writing, reviewing, and deploying software. Charity emphasizes that AI enables writing vast amounts of code quickly, transforming the economics of software engineering. She discusses the importance of verification, guardrails, and observability, stressing that as AI-generated code becomes more prevalent, ensuring reliability through testing, monitoring, and disciplined engineering practices becomes more critical. She also explores how systems are becoming non-deterministic, requiring new approaches to testing and validation.
Code Review and Observability in the Age of AI
A significant theme is the reconsideration of traditional artifacts like code reviews. Charity argues that as code becomes generated by AI, the process of review should shift from inspecting code to understanding behaviors, API designs, and system relationships. She advocates for better observability—especially instrumentation and telemetry—as foundational for managing non-determinism and ensuring system reliability. She explains how modern tools like open telemetry and telemetry-driven design are making instrumentation faster and more integrated into the development process.
AI's Role in DevOps and System Operations
The conversation touches on how AI changes DevOps practices. Charity reflects on the goal of creating tight feedback loops between development and production, noting that traditional DevOps has often failed in this regard. She discusses the importance of separation of concerns between infrastructure and code and the need for resilient, observable systems. She highlights recent trends in observability, automation, and how AI can both accelerate and complicate system operations.
Leadership and Management in an AI-Driven World
Charity emphasizes that modern leaders must be kind, skilled in business, and deeply involved in technical operations. She notes that with AI amplifying the pace of change, managers should return to hands-on roles and foster a culture of discipline, experimentation, and rapid iteration. The importance of managing change, owning responsibility, and maintaining trust with teams and vendors is stressed. She advocates for managers to be proactive, embrace AI skills, and validate their teams' work effectively.
The Future of Software Engineering and Human Skills
Key takeaways include that the value of human judgment, empathy, and context remains vital despite AI's capabilities. Charity advocates for continuous learning, emphasizing that engineers—especially juniors and even directors—should develop AI fluency to stay relevant. She underscores that mastery of domain, combined with ethical awareness and self-education, will determine future leadership and professional success. The importance of balancing AI automation with personal integrity and deep understanding is highlighted.
Sponsors
- Antisys · Antisys is verified software written by unreliable agents. It runs your whole system in a hostile simulation and roots out the bugs for you. It does this by using an approach called deterministic simulation testing, or DST, and this is a turbocharger testing by running a whole system under aggressive fault injection. Imagine, and this is as hundreds or thousands of versions of the Mario game running, each instance aggressively trying to break the game with increasingly weird input combinations. If it finds a breakage, this is where the determinism comes in. Instead of you having to try to reproduce a tricky bug you saw in production, this Antisys can provide you with a perfect deterministic replay of anything it finds every time. With Antisys, you can specify properties at the whole system level, and Antisys will actively try to disprove them. So you can be confident that if your system holds up in Antisys, it will hold up in production.
- WorkOS · WorkOS is built for how agents actually operate, with fine-grained authorization, SSO, and SCIM, not just user-auth with agents bolted on after. It ensures your app is enterprise-ready, providing the necessary security and management features.
- BuildKite · BuildKite is CI software designed to absorb whatever your coding agents throw at the build queue. It provides instant parallelization and unlimited concurrency to handle high volumes of commits, ensuring CI keeps up with rapid development. Trusted by major companies, it reliably scales to millions of job minutes weekly, capturing artifacts and logs instantly for quick insights.
Show notes
Brought to You By:
• Antithesis – verify your system’s correctness without human review or traditional integration tests – and avoid bugs or outages.
• WorkOS – everything you need to make your app enterprise ready.
• Buildkite – CI software built to absorb whatever your coding agents throw at the build queue
—
In 2025, it was rational to be skeptical about AI, but in 2026 it’s clear that AI is changing all of the industry, and there’s less and less place for skepticism. This take is from one of my favorite voices in software reliability and observability: Charity Majors, CTO and cofounder of Honeycomb, co-author of Observability Engineering. (Note: the second edition of Observability Engineering is out, and it’s pretty much a full rewrite of the book, I recommend grabbing it if you’re building reliable systems)
In this episode, I sat down with Charity to discuss how her thinking on AI has evolved, why she believes it is becoming a foundational part of software engineering, and what that means for how teams build, review, and ship software.
We explore how AI is changing the economics of code generation, why reliability and verification are increasingly the bottlenecks, and why the rise of non-deterministic systems requires more engineering discipline. Charity shares her views on code reviews, observability, DevOps, leadership, and why both AI skeptics and enthusiasts are getting important things right.
—
Timestamps
00:00 Intro
02:56 How Parse led to Honeycomb
06:00 The limits of individual productivity metrics
09:08 How Charity’s perspective on AI has evolved
13:50 Rewriting code vs. editing code
19:20 Production as a stage of development
22:14 Code reviews
26:56 Non-deterministic systems
31:11 Sensible uses of AI
37:41 The two AI camps
44:40 Why AI works so well for building software
49:42 DevOps
55:13 Modern observability
1:00:40 Handling context overload
1:01:56 What’s new in Observability Engineering’s 2nd edition
1:07:45 What effective leadership looks like
1:10:25 Engineering management: what is changing?
1:16:31 Junior engineers
1:18:01 AI fatigue
1:21:39 Book recommendations
—
The Pragmatic Engineer deepdives relevant for this episode:
• Deepdive: How 10 tech companies choose the next generation of dev tools
• Why is Meta destroying its engineering organization?
• When AI writes almost all code, what happens to software engineering?
• Are AI agents actually slowing us down?
• Observability: the present and future, with Charity Majors
• The third golden age of software engineering – thanks to AI, with Grady Booch
—
Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email podcast@pragmaticengineer.com.
Get full access to The Pragmatic Engineer at newsletter.pragmaticengineer.com/subscribe
Follow this show on AudioBrevity and new episodes get summarized for you. Create a free account