Only 50% of companies monitor their ML systems. Building observability for AI is not simple: it goes beyond 200 OK pings. In this episode, Sylvain Kalache sits down with Conor Brondsdon (Galileo) to unpack why observability, monitoring, and human feedback are the missing links to make large language model (LLM) reliable in production.

Conor dives into the shift from traditional test-driven development to evaluation-driven development, where metrics like context adherence, completeness, and action advancement replace binary pass-fail checks. He also shares how teams can blend human-in-the-loop feedback, automated guardrails, and small language models to keep AI accurate, compliant, and cost-efficient at scale.

Is “good code” still the right measure of engineering success in an AI-driven world? In this episode of Humans of Reliability, Rob Zuber, CircleCI CTO, joins Sylvain to explore how coding assistants are reshaping developer workflows and changing what teams value.

Rob shares what he’s seeing across CircleCI’s customer base: a clear boost in throughput, new bottlenecks shifting from code creation to code review, and the rise of “vibe coding,” where engineers trust AI-generated code they may not fully understand.

He challenges long-held assumptions about readability and maintainability, arguing that software engineering is on the edge of a paradigm shift. For SREs and developers alike, this conversation is a candid look at how to stay relevant, embrace simplicity, and rethink reliability in the age of AI.

What does it really take to move from firefighting incidents to building reliability at scale? In this episode of Humans of Reliability, Shery Brauner (Razor, ex-Zalando) shares her unique journey from frontend and backend engineering to leading site reliability practices. She explains why protecting the user journey is the key to effective incident management, how SLOs cut through noisy alerts, and why observability must come first.

Shery also talks about practical steps teams can take to adopt an SLO-driven strategy, the pitfalls of over-instrumentation, and how AI is shaping the future of incident response. Whether you’re an engineer, manager, or reliability leader, you’ll walk away with concrete ideas on how to protect what matters most: your customers’ experience.

Sylvain Kalache sits down with Brian Shaw, Senior Engineering Leader at Uphold, to explore the reliability challenges that arise when operating at the intersection of traditional finance and crypto markets.

Brian shares how unexpected market events can create massive traffic spikes, how their platform architecture and Kubernetes setup help them stay resilient, and why Uphold's transparency and regulatory approach make them both trustworthy and a high-profile target.

The conversation also touches on AI's emerging role in operations, lessons from major incidents, and the delicate balance between innovation and stringent compliance in financial engineering.

Incident response is as much about people as it is about systems. In this episode, David Owczarek, a veteran engineer leader and seasoned incident commander, joins Silvan Kalache to unpack the human dynamics behind effective reliability leadership.

Drawing on experiences across startups and global enterprises, David shares what really matters when everything breaks, including:

– How incident response strategies shift between small companies and large enterprises

– Why not every engineer should be an incident commander

– How empathy and transparency during outages can deepen customer trust instead of eroding it

– Where AI is showing promise in SRE workflows—and where human judgment still reigns supreme

If you’re leading incidents, training ICs, or thinking about the future of AI-assisted reliability, this episode is packed with hard-won insights and grounded strategies for navigating chaos with clarity.

AI is transforming reliability work—from reactive firefighting to proactive engineering. In this episode, Ryan Lockard, VP of Platform Engineering and AI Enablement at CVS Health, joins Sylvain Kalache to break down how AI is showing up on the frontlines of healthcare infrastructure and operations.

From LLM copilots to cultural shifts in ownership, Ryan walks us through:

• How AI tools help troubleshoot legacy systems and assist during real-time incidents
• Why proactive reliability is finally within reach thanks to AI-enhanced tooling and workflows
• What MCP servers are, and how natural language interfaces are streamlining cloud operations
• How engineering culture and on-call models shift when teams truly own their reliability posture
• What AI means for the next generation of developers—and why prompt engineering matters

Whether you're managing incidents, building platforms, or scaling team culture, this conversation offers a pragmatic lens on how AI is changing the work of reliability from the inside out.

In this episode, we sit down with Cosmo Wolfe, Head of Technology at Metronome, to unpack how reliability, trust, and architecture intersect in one of the most critical and overlooked parts of the AI product stack: billing.

As AI workloads introduce unpredictable usage patterns and nontraditional pricing models—from token-based to outcome-based—companies are navigating a new frontier of customer trust. Cosmo explains why billing is more than just a backend function; it’s a key moment of truth in the product experience.

We explore how event-sourced systems, rigorous monitoring, and internal accountability help avoid trust-eroding mistakes, like misbilled invoices or opaque usage tracking. Whether it’s ensuring a usage cap actually activates or being able to reconstruct billing history for an enterprise customer, this episode reveals how reliability at the billing layer becomes a strategic advantage in a competitive, AI-driven software landscape.

Internal platforms promise speed, consistency, and scale — but what happens when they become a distraction? In this episode, Chase Roberts, COO at Northflank, joins Sylvain Kalache to examine the quiet ways platforms erode developer experience when not planned carefully. 

From abandoned golden paths to shadow deployments and brittle YAML pipelines, Chase walks us through: 

• Why early PaaS got developer experience right and what it missed 
• The cultural bias toward building over buying (and its hidden costs) 
• How complexity quietly kills productivity and reliability, even when everything “works” 
• The three questions every team should ask before building an IDP 
• What a future of truly portable, resilient platforms might look like 

Episode links: 

• Build vs. buy: The platform engineer’s conundrum

In this episode of Humans of Reliability, we sit down with Justin Reock, Deputy CTO at DX, to unpack the real impact of generative AI on developer productivity. Drawing from early data in DX’s GenAI Impact Report, he explains why time savings alone don’t tell the full story and why the real value might lie in shifting cognitive load toward meaningful work. 

We also explore how traditional productivity metrics like PR throughput can backfire, why teams need to move beyond DORA, and how modern frameworks like SPACE and the DX Core 4 offer a more complete view of engineering health. 

Episode links:

• Guide to AI assisted engineering
• DX Core 4

In this episode, we explore how networking has shaped reliability as we know it. Marino Wijay cloud networking expert and Staff Solutions Architect at Kong shares how his journey began not as an SRE, but with cables, routers, and switches.

Marino explains the evolution of the fabric holding systems together through virtualization, and how software-defined networking, which is now a key element to resilient applications.

This episode also dives into the new challenges LLMs are introducing into networking. Marino discusses how these workloads introduce new types of reliability challenges: longer response times, context preservation, model switching, and request sanitization.

Finally, Marino emphasizes the human role. Behind every protocol, abstraction layer, and model prompt is a human providing context, compassion, and a sense of accountability.