Super Data Science: ML & AI Podcast with Jon Krohn 958: Without Trusted Context, Agents are Stupid (featuring Salesforce’s Rahul Auradkar)
32 snips
Jan 16, 2026 In this engaging discussion, Rahul Auradkar, EVP at Salesforce, dives into the importance of unified data engines, particularly following the company’s acquisition of Informatica. He explains how Data 360 facilitates actionable insights, while contrasting the concepts of context versus data in AI models. Discover how Tableau enhances analytics and why trusted context is critical for intelligent agent behavior. Rahul also shares insights on upskilling teams with no-code tools and envisions the future of agent-driven workflows by 2026.
AI Snips
Chapters
Books
Transcript
Episode notes
Data Rich But Context Poor
- Enterprises are often data rich but context poor, which makes AI models 'corporate stupid'.
- Providing unified, harmonized customer context grounds agents and enables useful, consistent automation.
Tableau Enables Agentic Analytics
- Tableau supplies an intuitive analytics experience and a semantic layer that enables agentic, natural-language exploration.
- Combining Tableau with unified metadata makes analytics actionable and agent-ready.
Integration Layer Powers Agent Actions
- MuleSoft provides app-to-app integration, API governance, and an 'agent fabric' for orchestrating multi-agent actions.
- That integration layer turns unified data into actionable automations at endpoints.
