Snowflake Hosts CTO Summit on Building AI-Native Engineering Teams

Snowflake held the inaugural CTO Circle during its 2026 Summit in San Francisco, bringing together more than 350 CTOs from financial services, telecommunications, retail, and technology. The event was designed to give engineering leaders a trusted forum to learn from peers navigating the transformation to AI-native organizations and to challenge assumptions—a conversation rarely shared openly because it remains shaped by competitive pressures and rapidly evolving technology.

Vivek Raghunathan, SVP of Engineering at Snowflake, framed the transformation as a management philosophy shift. Most organizations begin by introducing coding assistants, which deliver meaningful but incremental improvements. Snowflake instead treated developer productivity as a product problem. "What if you treated your developers like customers?" became the foundation for its engineering transformation. The team interviewed developers to understand where work slowed down, mapped friction points across the software development lifecycle, established baseline metrics, and ran experiments to gauge the impact of every change. This bottom-up approach, combined with clear executive sponsorship, ensured that improvements reflected how engineers actually worked rather than how leaders expected them to work. The results were measurable: in 18 months, Snowflake increased its internal developer Net Promoter Score by more than 30 points, achieving a 4:1 ratio of satisfied to dissatisfied developers. More importantly, that improvement translated into an engineering organization capable of delivering software more efficiently and adapting more quickly as AI capabilities evolved.

Raghunathan explained that adoption of AI tools alone is not enough; the real leverage comes from depth of usage and true mastery at scale. This journey evolves through three stages: adoption (when developers learn to use AI tools daily), mastery (when engineers discover repeatable workflows that consistently produce better outcomes), and optimization (when those workflows become organizational knowledge available to every engineer). At Snowflake, early AI adopters experimented with prompting techniques, planning methods, and debugging approaches; over time, the organization documented patterns that delivered better results and made them available across engineering. Every developer had access to the same tools, but engineers using proven workflows consistently outperformed those stuck at early adoption levels. This emphasis on workflows also changes how organizations think about software development itself. As AI reduces the effort to transform ideas into working software, everyone becomes a builder—product managers can prototype experiences, designers can validate concepts directly in code, and domain experts can add guardrails. Code increasingly becomes the fastest way to test assumptions.

Jon McNeill, author of The Algorithm, expanded on this by encouraging leaders to reconsider decades-old assumptions. Too often, companies begin with technology and search for places to apply it; successful organizations reverse that equation by identifying the few business constraints that matter most and redesigning engineering systems around solving those problems. As organizations move AI into production, success will be measured not by who generates the most code or consumes the most tokens, but by who builds the simplest, fastest, and most effective engineering system.

Jeremy Burton, General Manager of the Observability Business Unit at Snowflake, addressed the second major theme: balancing velocity with risk. Early experimentation is giving way to production deployment, and organizations are increasingly expected to demonstrate measurable business value rather than isolated technical successes. AI introduces entirely new operational challenges—AI agents generate more telemetry, interact with more systems, and make decisions using information distributed across increasingly complex environments. Burton challenged a common assumption that better models simply need better data; in reality, AI is only as effective as the context it can access. That context extends far beyond raw telemetry to include semantics describing what data means, relationships captured through ontologies and knowledge graphs, and business context connecting systems. AI agents need standardized interfaces—APIs, CLIs, and Model Context Protocol (MCP)—to reliably retrieve and act on information, while engineers need intuitive ways to drill into data and validate AI-generated insights.

Aditya Gaur, Engineering Manager at Netflix, demonstrated this in practice through automated root cause analysis. Although often described as an AI initiative, its success depended far more on data architecture. Years before introducing AI agents, Netflix invested in connecting fragmented telemetry across systems, modeling operational relationships through an ontology and knowledge graph, and creating a shared context layer. By the time AI entered the picture, the foundation already existed; AI agents could reason over structured operational knowledge and generate far more meaningful hypotheses during incident investigations rather than searching across disconnected logs and dashboards.

Caitlin Colgrove, CTO at Hex, approached velocity from an organizational perspective. She urged leaders to rethink what it means to move fast in the AI era—organizations cannot afford to be partially AI native and eventually reach a point where they must commit fully by restructuring how teams build products, make decisions, and incorporate AI into everyday work. Hex initially created a dedicated AI product team, which delivered useful features but created an organizational bottleneck. The company ultimately disbanded the centralized AI organization and distributed responsibility across every product team. Today, Hex delivers product rapidly and embeds AI capabilities throughout the product because ownership lives with engineers closest to the customer problem.

Chris Kozlowski, Managing Director at Barclays, brought an enterprise perspective, highlighting that speed only creates value when matched with governance and trust. In highly regulated environments, engineering teams cannot afford to choose between innovation and control. Corey Burke, SVP Engineering at Dialpad, and Arun Rajamanickam, VP of Engineering at project44, described how AI is changing the pace of software development—as AI agents become capable of implementing significant portions of a feature independently, engineers spend less time writing code and more time defining intent, orchestrating multiple agents, and validating outcomes. Maximizing velocity requires building engineering platforms that allow teams to experiment quickly without compromising reliability.

The discussion also touched on how AI inevitably changes engineering organization design. Qi Jin, EVP at Cerebras Systems, suggested that organizations optimized for previous generations of software development often become the biggest barrier to AI adoption, as established processes and boundaries were designed to scale proven ways of working, not to continuously adapt.