CTO Circle Convenes 350 Engineering Leaders on Building AI-Native Organizations

Snowflake held the inaugural CTO Circle event during Snowflake Summit 2026 in San Francisco, gathering more than 350 CTOs from financial services, telecommunications, retail, and technology sectors. The event was created to give engineering leaders a trusted forum to learn from peers navigating the same transformation and to challenge assumptions about building AI-native organizations—a topic for which few established playbooks exist.

Vivek Raghunathan, SVP of Engineering at Snowflake, opened the discussion by challenging a widely held assumption: that developer productivity is an engineering or culture problem. Instead, Snowflake began treating developer productivity as a product. The team applied the same product management principles used to build customer-facing products: they interviewed developers to identify friction points, mapped the software development lifecycle, established baseline metrics, and ran experiments to measure the impact of every change. This approach combined clear executive sponsorship with bottom-up adoption, ensuring 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.

Raghunathan explained that adoption alone is insufficient; competitive advantage comes from depth of usage and mastery at scale. He outlined three stages: adoption (developers learning to use AI tools), mastery (engineers discovering repeatable workflows that deliver better outcomes), and optimization (workflows becoming organizational knowledge accessible to all engineers). At Snowflake, this manifested in collecting engineering design patterns developed by early AI adopters—experimenting with prompting techniques, planning methods, and debugging approaches—and making those patterns available across the organization. Engineers using proven workflows consistently outperformed those at early adoption stages, even though all had access to the same tools. This shift changes how organizations think about software development itself: as AI reduces the effort to transform ideas into working software, product managers can prototype experiences, designers can validate concepts in code, and domain experts can add guardrails. Code becomes the fastest way to test assumptions.

Jon McNeill, author of The Algorithm, urged leaders to reverse conventional thinking: instead of beginning with technology and searching for applications, successful organizations identify the few business constraints that matter most and redesign engineering systems around solving those problems. Success in moving AI to production will be measured less by who generates the most code or consumes the most tokens and more by who builds the simplest, fastest, and most effective engineering system.

Jeremy Burton, General Manager of the Observability Business Unit at Snowflake, addressed a second challenge: maximizing velocity while containing risk. He argued that AI has entered a new phase where early experimentation is giving way to production deployment, and organizations must demonstrate measurable business value. AI introduces new operational complexity: agents generate more telemetry, interact with more systems, and make decisions using distributed information. Jeremy challenged the common assumption that better models simply need better data. In reality, AI is only as effective as the context it can access—context that extends far beyond raw telemetry to include semantics, ontologies, knowledge graphs, and business context. AI agents need standardized interfaces such as APIs, CLIs, and Model Context Protocol (MCP) to reliably retrieve and act on information. Rich context combined with open access transforms disconnected telemetry into an environment where humans and AI can reason effectively and make better decisions.

Aditya Gaur, Engineering Manager at Netflix, illustrated this principle through Netflix's automated root cause analysis project. Although described as an AI initiative, its success depended far more on data architecture. Netflix invested years before deploying AI agents in connecting fragmented telemetry, modeling operational relationships through an ontology and knowledge graph, and creating a shared context layer. When AI entered the picture, agents could reason over structured operational knowledge rather than searching disconnected logs, generating far more meaningful hypotheses during incident investigations.

Caitlin Colgrove, CTO at Hex, approached velocity from an organizational perspective, urging leaders to rethink what moving fast means in the AI era. Organizations cannot remain partially AI-native; they must eventually commit fully by restructuring how teams build products, make decisions, and incorporate AI into everyday work. Hex initially created a dedicated AI product team, but this approach created an organizational bottleneck. The company disbanded the centralized AI organization and distributed responsibility across every product team, allowing them to deliver products rapidly and embed AI capabilities throughout because ownership lived with engineers closest to customer problems.

Chris Kozlowski, Managing Director at Barclays, emphasized that speed only creates value when matched with governance and trust, particularly in highly regulated environments where engineering teams cannot 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 features independently, engineers spend less time writing code and more time defining intent, orchestrating agents, and validating outcomes. They emphasized that maximizing velocity requires platforms that allow teams to experiment quickly without compromising reliability.

Qi Jin, EVP at Cerebras Systems, suggested that organizations optimized for previous software development generations often become the biggest barrier to AI adoption. Established processes and organizational boundaries were designed to scale proven ways of working, not to continuously adapt as capabilities evolve. The implication is that true AI-native organizations must be willing to restructure fundamentally, not merely bolt AI onto existing engineering systems.