THE THYNK MANIFESTO

Speed and agility,
at the venue's scale.

In December 2025, Sam Altman publicly admitted his concern about the rate of change that's happening in the world right now. He was speaking on a late-night talk show. The line stuck.

The convention-centre commercial team feels it. Cvent windows shrank from a week to twelve hours. Exhibitors expect Amazon-grade ordering. Operations teams expect mobile-first workflows. Finance teams expect SAP-grade integration. Procurement teams now ask not "do you have AI?" but "where does your data live, who runs your AI, and what does it touch when it acts?"

The fact pattern is no longer that AI is coming. The fact pattern is that the rate of organisational change required to use AI well is the new commercial bottleneck. Speed of safe change becomes the moat.

In March 2026, the venture investor George Sivulka argued that AI has made individual workers ten times more productive but has not made companies ten times more valuable. His reason: organisations have not been redesigned to absorb it. Most AI in the wild is Individual AI — chat assistants, prompts, fragmented productivity. Institutional AI is something else. Institutional AI coordinates. It filters signal. It produces objectivity. It empowers everyone the venue depends on. It acts on what nobody asked.

We agree.

Thynk is not a chat assistant. Thynk is the layer underneath the chat assistants — the institutional data layer, the institutional workflow layer, the institutional domain ontology, the institutional memory of an industry we have spent fifteen years inside.

We are built on Salesforce because Salesforce is the world's most documented institutional data platform. We ship five portals because no venue commercial institution operates on one role's interface. We publish 212 objects and 4,173+ fields because no institution can govern AI it cannot inspect. We hire the people who built the system you are leaving because institutional memory is the deepest moat in software.

We believe what we ship matters most when our customers stop noticing it. When the inbound RFP routes itself to the right rep, drafts itself with the right rates, opens the booking with the right BEOs, alerts the supplier with the right delivery window — and nobody had to prompt any of it — the venue is operating at the speed Altman warned us about.

That is what institutional AI for venues looks like.

We will keep shipping.

"We have our electricity.
It's time to redesign our factories." — George Sivulka, a16z, March 2026