CXOTalk Advice on Agentic AI: A Top Snowflake Exec Explains | CXOTalk #903
Dec 27, 2025
Christian Kleinerman, EVP of Product at Snowflake, with 25 years of expertise in enterprise tech, dives deep into the readiness for agentic AI. He emphasizes data organization and governance as key success factors. Kleinerman shares how to evaluate AI vendors efficiently and highlights practical use cases, like coding assistants that enhance productivity across roles. He advocates for rapid experimentation, cautioning against risks like incorrect outputs and security issues. With pragmatic advice, he provides a roadmap for enterprises venturing into AI.
AI Snips
Chapters
Transcript
Episode notes
Context Beats Generic Knowledge
- Agents unlock far more value when they access enterprise-specific context rather than generic internet knowledge.
- Christian Kleinerman says contextualized agents enable retrieval and action like closing tickets or drafting customer replies.
Roll Out Small, Scale On Success
- Start with small, well-scoped rollouts and scale only after proving success in production.
- Christian Kleinerman advises constant iteration and frequent re-evaluation as models rapidly improve.
Correctness And Governance Are Non-Negotiable
- Two common failure causes are incorrect outputs and improper access to sensitive data.
- Kleinerman emphasizes solutions must deliver correctness while enforcing governance and security.