Description
AI agents are only as useful as the data they can reach, and most company data is scattered across clouds, data centers, and legacy systems. Giving agents access is easy to get wrong, especially when the data includes salaries, customer records, or financial results. That's why governance, permissions, and audit trails are becoming the main bottleneck for getting agents into production. How do you give an agent the right access without exposing sensitive information? Who owns that job? And do traditional ideas like data fabrics and data meshes still hold up?
Sergio Gago is the CTO at Cloudera. He previously held senior roles at Moody's Analytics, most recently as managing director of AI/ML and quantum, and was CTO of Rakuten's Spanish operations and of digital publishing company Zinio.
In the episode, Richie and Sergio explore hybrid infrastructure for AI, choosing between cloud, neo clouds and on-premise, data fabrics and governance for agents, why agents need data access, the changing role of data teams, ephemeral dashboards, skills for engineers, sovereign AI, and much more.
0:00 Intro
1:15 Has AI reversed the move to the cloud?
3:22 Cloud, neo cloud or on-prem: how to decide
9:29 Managing messy infrastructure once agents arrive
15:15 Data fabric vs data mesh: what works now
17:42 Why agent permissions get so complicated
21:22 Bring AI to the data, not data to AI
29:42 What the data team does now
31:16 Ephemeral dashboards and agent-built analytics
35:53 Skills for data and software engineers
38:48 New infrastructure jobs and the hybrid boom
41:02 Orchestrating compute across clouds and data centers
44:35 Sergio's UN talk on AI sovereignty
Find DataFramed on DataCamp https://www.datacamp.com/podcast and on your preferred podcast streaming platform:
Apple Podcasts: https://podcasts.apple.com/us/podcast/dataframed/id1336150688
Spotify: https://open.spotify.com/show/02yJXEJAJiQ0Vm2AO9Xj6X?si=d08431f59edc4ccd
Links Mentioned in the Show:
• Cloudera: The Great AI Rearchitecture report - https://www.cloudera.com/about/news-and-blogs/press-releases/2026-08-11-ninety-five-percent-of-enterprises-have-delayed-ai-projects-as-infrastructure-limitations-spark-the-great-ai-re-architecture.html
• Thoughtworks: Core principles of data mesh - https://www.thoughtworks.com/en-gb/about-us/events/webinars/core-principles-of-data-mesh
• Cloudera Unified Data Fabric - https://www.cloudera.com/content/dam/www/marketing/resources/datasheets/accelerate-ai-with-a-unified-data-fabric.pdf
• Apache Ranger - https://ranger.apache.org/
• OpenClaw explained - https://operator.io/blog/what-is-openclaw
• Trino - https://trino.io/
• Connect with Sergio: [LinkedIn URL — to confirm]
• AI-Native Course: Intro to AI for Work - https://www.datacamp.com/courses/introduction-to-ai-for-work
• Related Episode: AI Agents Are the New Shadow IT (And Your Governance Isn't Ready) with Stijn Christiaens - https://www.datacamp.com/podcast/ai-agents-are-the-new-shadow-it
New to DataCamp? Learn on the go using the DataCamp mobile app - https://www.datacamp.com/mobile
Empower your business with world-class data and AI skills with DataCamp for business - https://www.datacamp.com/business