In this interview, Shubhangi Dua, Podcast Host, Producer and B2B Tech Journalist at EM360Tech, speaks to Jane Smith, Field Chief Data & AI Officer for EMEA at ThoughtSpot, at Big Data LDN 2026 at Kensington Olympia, London.
They talk about how the industry has moved from the theoretical promise of "Agentic" AI to practical implementation, focusing on context, semantic layers, and managing token costs.
“Last year was about the promise of agentic. It was theoretical and high-level, but this year, it’s here,” Smith tells Dua. 2026 is about the “implementation of Agentic, the practicalities and the context, semantic layers and the token costs.”
The mindset is shifting to enterprise economics. “The enterprise economics of token costs, also referred to as tokenomics, the economics of engineering costs to support the token costs,” Smith adds. The bill has come due, and people are reflecting on the past when they claimed how AI would save them money.
“Tokenmaxxing is dead,” the Field Chief Data and AI Officer says. There are many ways to optimise for tokens. Enterprises should build proper architecture and infrastructure first. Additionally, they shouldn’t bypass good engineering principles.
“Don’t do context stuffing. Offload to MCP where you can,” she advises. “If you're on your structured data, you don't necessarily need to have text-to-SQL generated; offload it to a deterministic tool.
Visit thoughtspot.com.
Also Read: Agentic AI: ThoughtSpot's Spotter AI is Making Businesses Autonomous
