DevOps Paradox

Matt DeBergalis, CEO of Apollo GraphQL and longtime GraphQL practitioner, argues that AI agents are turning internal APIs into the customer-facing interface. He explains why typical REST APIs overload models and how GraphQL can orchestrate concise queries to reduce hallucination, cost, and latency. He also covers permission models, declarative infrastructure for many ephemeral APIs, and why docs must be written for models first.

They dig into why most corporate docs are hard to find, out of date, and mismatch reality. They contrast eternal 'why' notes with fragile 'how' details that decay the moment configs change. They explain where truth lives today — live systems, meeting transcripts, and people's heads. They argue RAG alone won't help and outline practical steps: audit, instrument, design for retrieval, and keep indexing continuously.

Trevor Stuart, co-founder of Split.io and head of Feature Management & Experimentation at Harness, has deep experience with feature flags and experimentation. He discusses AI-written code creating a six-lane highway into a two-lane bridge of reviews and delivery. Teams are embedding AI configs into flags and A/B testing agents in production. Culture, flag lifecycle, and running revenue-driving experiments receive special focus.

A discussion about why operations teams prefer stability and often resist new waves like cloud, containers, and Kubernetes. The hosts explore how tool origins and tribal identity shape adoption decisions. They highlight shadow IT as a signal for change and argue for incremental migrations. They also cover practical AI uses for ops and offer ways to persuade cautious teams without promising perfection.

Anthony Eden, Founder and CEO of DNSimple with decades in DNS and domain registration. He explains why decades-old DNS still underpins everything online. Topics include running DNS at scale, unicast vs anycast, resolver proximity, common domain mistakes like running production DNS on registrars, email deliverability woes, and practical AI uses for engineers.

They explore why adding AI to coding speeds one step but shifts the bottleneck downstream to QA, security, legal and others. They argue the fastest flow is a single person owning idea-to-production with automation for tests and deploys. They contrast true automation with simply hiring people to do tasks and warn that AI is a systems problem that forces organization-wide change.

Evan Kaplan, CEO of InfluxData and longtime leader in time series databases, discusses real-time data for physical automation. He talks about why nanosecond-accurate sampling and low-latency storage matter for robotics and autonomous vehicles. He explains trade-offs in sampling and storage, the need for deterministic models in safety-critical systems, and how open source licensing and cloud partnerships are evolving.

A provocative look at a future where employees bring their own AI teams into the workplace. Short scenes imagine interviewing someone who expects to use personal AI agents. Legal, compensation and intellectual property headaches from AI-trained stacks are examined. The hosts compare this shift to BYOD and debate how organizations must adapt or lose top talent.

Peter Zaitsev, co-founder of Coroot and observability entrepreneur, talks about opinionated, self-hosted monitoring that favors simplicity. He covers automatic instrumentation with eBPF, deterministic root-cause analysis, AI-assisted recommendations, and how Coroot fits alongside other tools. The conversation also touches licensing, bootstrapped funding, deployment options, and tradeoffs in telemetry and retention.

Explore how AI is reshaping the role of developers, highlighting that coding itself is the easy part. Discover the breakdown of traditional roles as AI fosters a return to pair programming dynamics. Communication skills become crucial, as developers must connect technical decisions with non-technical stakeholders. Juniors have a unique chance to build foundational knowledge while adapting rapidly. The future sees seniors acting as editors of AI-generated work, emphasizing adaptability and curiosity as keys to success in a changing landscape.