Share Episode ⸺ Episode Sponsor: Incident.io - https://dev0ps.fyi/incidentio

Warren is joined by Olga Kundzich, Co-founder and CTO of Moderne, to discuss the reality of technical debt in modern software engineering. Olga reveals a shocking statistic: without maintenance, cloud-native applications often cease to function within just six months. And from our experience, that's actually optimistic. The rapid decay isn't always due to bad code choices, but rather the shifting sands of third-party dependencies, which make up 80 to 90% of cloud-native environments.

We review the limitations of traditional Abstract Syntax Trees (ASTs) and the introduction of OpenRewrite's Lossless Semantic Trees (LSTs). Unlike standard tools, LSTs preserve formatting and style, allowing for automated, horizontal scaling of code maintenance across millions of lines of code. This fits perfectly in to the toolchain that is the LLMs and open source ecosystem. Olga explains how this technology enables enterprises to migrate frameworks—like moving from Spring Boot 1 to 2 — without dedicating entire years to manual updates.

Finally, they explore the intersection of AI and code maintenance, noting that while LLMs are great at generating code, they often struggle with refactoring and optimizing existing codebases. We highlight that agents are not yet fully autonomous and will always require "right-sized" data to function effectively. Will is absent for this episode, leaving Warren to navigate the complexities of mass-scale code remediation solo.

• DevOps Episode: We read code
• DevOps Episode: Dynamic PRs from incidents
• OpenRewrite
• Larger Context Windows are not better

• Warren - Dell XPS 13 9380
• Olga - Claude Code

Share Episode

Microsoft's John Papa, Partner General Manager of Developer Relations for all things dev and code joins the show to talk developer relations...from his Mac. He reveals his small part in the birth of VS Code (back when its codename was Ticino) after he spent a year trying a new editor every month.

The conversation dives deep into "Agentic AI," where John predicts developers will soon become "managers of agents". But is it all hype? John and Warren debate the risks of too much automation (no, AI should not auto-merge your PRs) and the terrifying story of a SaaS built with "zero handwritten code" that immediately got hacked because the founder was "not technical".

The episode highlights John's jaw-dropping war stories from Disney, including a mission-critical hotel lock system (for 5,000+ rooms) that was running on a single MS Access database under a desk. It's a perfect, cringeworthy lesson in why "we don't have time to test" is the most expensive phrase in tech, and why we need a human in the loop. John leaves us with the one question we must ask of all new AI features: "Who asked for that?"

• Impact of AI on Critical Thinking paper
• LLMs raise the floor not the ceiling
• DevOps Episode: How far along with AI are we?

• Warren - Shokz OpenFit 2
• John - Run Disney

Share Episode ⸺ Episode Sponsor: Attribute - https://dev0ps.fyi/attribute

In the wake of one of the worst AWS incidents in history, we're joined by Lawrence Jones, Founding Engineer at Incident.io. The conversation focuses on the challenges of managing incidents in highly regulated environments like FinTech, where the penalties for downtime are harsh and require a high level of rigor and discipline in the response process. Lawrence details the company's evolution, from running a monolithic Go binary on Heroku to moving to a more secure, robust setup in GCP, prioritizing the use of native security primitives like GCP Secret Manager and Kubernetes to meet the obligations of their growing customer base.

We spotlight exactly how a system can crawl GitHub pull requests, Slack channels, telemetry data, and past incident post-mortems to dynamically generate an ephemeral runbook for the current incident.Also discussed are the technical challenges of using RAG (Retrieval-Augmented Generation), noting that they rely heavily on pre-processing data with tags and a service catalog rather than relying solely on less consistent vector embeddings to ensure fast, accurate search results during a crisis.

Finally, Lawrence stresses that frontier models are no longer the limiting factor in building these complex systems; rather, success hinges on building structured, modular systems, and doing the hard work of defining objective metrics for improvement.

• Cloud Secrets management at scale
• Episode: Solving Time Travel in RAG Databases
• Episode: Does RAG Replace keyword search?

• Warren - Anker Adpatable Wall-Charger - PowerPort Atom III
• Lawrence - Rocktopus & The Checklist Manifesto

Share Episode ⸺ Episode Sponsor: Attribute - https://dev0ps.fyi/attribute

We're joined by 20 year industry veteran and DevOps advocate, Adam Korga, celebrating the release of his book IT Dictionary. In this episode we quickly get down to the inspiration behind postmortems as we review some cornerstone cases both in software and in general technology.

Adam shares how he started in the industry, long before DevOps was a coined term, focused on making systems safer and avoiding mistakes like accidentally dropping a production database. we review the infamous incidents of accidental database deletion, by LLMs and human's alike.

And of course we touch on the quintessential postmortems in civil engineering, flight, and survivorship bias from World War II through analyzing bullet holes on returning planes.

• Adam's book: IT Dictionary
• Knight Capital: the 45 minute nightmare
• Work Chronicles Comic: Will my architecture work for 1 Million users?

• Warren - Cuitisan CANDL storage containers
• Adam - FUBAR

• Episode: MCP: The Model Context Protocol and Agent Interactions
• Crossing the Chasm

• Warren - HanCenDa USB C Magnetic adapter
• Jenna - Keychron Alice Layout Mechanical keyboard (And get a 5% discount on us)

Share Episode

This episode we are joined by Andrew Moreland, co-founder of Chalk. Andrew explains how their company's core business model is to deploy their software directly into their customers' cloud environments. This decision was driven by the need to handle highly sensitive data, like PII and financial records, that customers don't want to hand over to a third-party startup.

The conversation delves into the surprising and complex challenges of this approach, which include managing granular IAM permissions and dealing with hidden global policies that can block their application. Andrew and Warren also discuss the real-world network congestion issues that affect cross-cloud traffic, a problem they've encountered multiple times. Andrew shares Chalk's mature philosophy on software releases, where they prioritize backwards compatibility to prevent customer churn, which is a key learning from a competitor.

Finally, the episode explores the advanced technical solutions Chalk has built, such as their unique approach to "bitemporal modeling" to prevent training bias in machine learning datasets. As well as, the decision to move from Python to C++ and Rust for performance, using a symbolic interpreter to execute customer code written in Python without a Python runtime. The episode concludes with picks, including a surprisingly popular hobby and a unique take on high-quality chocolate.

• Fact - The $1M hidden Kubernetes spend
• Giraffe and Medical Ruler training data bias
• SOLID principles don't produce better code?
• Veritasium - The Hole at the Bottom of Math
• Episode: Auth Showdown on backwards compatible changes

• Warren - Switzerland Grocery Store Chocolate
• Andrew - Trek E-Bikes

Share Episode

We welcome guest Ang Li and dive into the immense challenge of observability at scale, where some customers are generating petabytes of data per day. Ang explains that instead of building a database from scratch—a decision he says went "against all the instincts" of a founding engineer—Observe chose to build its platform on top of Snowflake, leveraging its separation of compute and storage on EC2 and S3.

The discussion delves into the technical stack and architectural decisions, including the use of Kafka to absorb large bursts of incoming customer data and smooth it out for Snowflake's batch-based engine. Ang notes this choice was also strategic for avoiding tight coupling with a single cloud provider like AWS Kinesis, which would hinder future multi-cloud deployments on GCP or Azure. The discussion also covers their unique pricing model, which avoids surprising customers with high bills by providing a lower cost for data ingestion and then using a usage-based model for queries. This is contrasted with Warren's experience with his company's user-based pricing, which can lead to negative customer experiences when limits are exceeded.

The episode also explores Observe's "love-hate relationship" with Snowflake, as Observe's usage accounts for over 2% of Snowflake's compute, which has helped them discover a lot of bugs but also caused sleepless nights for Snowflake's on-call engineers. Ang discusses hedging their bets for the future by leveraging open data formats like Iceberg, which can be stored directly in customer S3 buckets to enable true data ownership and portability. The episode concludes with a deep dive into the security challenges of providing multi-account access to customer data using IAM trust policies, and a look at the personal picks from the hosts.

• Fact - Passkeys: Phishing on Google's own domain and It isn't even new
• Episode: All About OTEL
• Episode: Self Healing Systems

• Warren - The Shadow (1994 film)
• Ang - XREAL Pro AR Glasses

Share Episode

In a special solo flight, Warren welcomes Meagan Cojocar, General Manager at Pulumi and a self-proclaimed graduate of “PM school” at AWS. They dive into what it's like to own an entire product line and why giving up that startup hustle for the big leagues sometimes means you miss the direct signal from your users. The conversation goes deep on the paradox of open-source where direct feedback is gold, but dealing with license-shifting competitors can make you wary. From the notorious HashiCorp kerfuffle to the rise of OpenTofu, they explore how Pulumi maintains its commitment to the community amidst a wave of customer distrust.

Meagan highlights the invaluable feedback loop provided by the community, allowing for direct interaction between users and the engineering team. This contrasts with the "telephone game" that can happen in proprietary product development. The conversation also addresses the recent industry shift and then immediate back-peddling from open-source licenses, discussing the subsequent customer distrust and how Pulumi maintains its commitment to the open-source model.

And finally, the duo tackles the elephant in the cloud: LLMs, and extends on the earlier MCP episode. They debate the great code quality vs. speed trade-off, the risk of a "botched" infrastructure deployment, and whether these models can solve anything more than a glorified statistical guessing game. It's a candid look at the future of DevOps, where the real chaos isn't the code, but the tools that write it. The conversation concludes with a philosophical debate on the fundamental capabilities of LLMs, questioning whether they can truly solve "hard problems" or are merely powerful statistical next-word predictors.

• Veritasium - the Math that predicts everything
• Fact - Don't outsource your customer support: Clorox sues Cognizant
• CloudFlare uses an LLM to generate an OAuth2 Library

• Warren - Rands Leadership Community
• Meagan - The Manager's Path by Camille Fournier