Kuber Sharma is Senior Director of Product Marketing at UiPath, where he leads go-to-market for agentic automation. Twelve years in enterprise software across Microsoft Azure, Salesforce, Tableau and UiPath, and twenty years in marketing since starting out in Delhi media.
kubersharma.com · Positioned on Substack · LinkedIn · kuber@kubersharma.com
Four models built out of real enterprise launches, not a whiteboard. Full write-ups live on kubersharma.com. Each one is linked below.
The Augmented Marketing Decision Architecture (AMDA)
Three zones for allocating marketing and go-to-market decisions between AI and humans, sorted by whether a decision is reversible, who sees the output, and how often it happens. AI-Primary, Collaborative, Human-Primary. Developed at Tableau, a Salesforce company, and applied since at UiPath. Runnable as a Claude skill: amda-skill. Reference card.
Five stages explaining why enterprise AI succeeds in a demo and dies before production. Excitement, Scoping, Sandboxing, Stall, Abandonment. The trap springs at stage three, where the pilot works beautifully in a sandbox that has none of the integrations, none of the policies and none of the people who own production. Names the six infrastructure gaps a pilot was never designed to close. Reference card.
Three layers of trust an enterprise buyer needs before deploying AI at scale. Technical Trust, then Operational Trust, then Strategic Trust. Buyers evaluate bottom to top. Vendors pitch top to bottom. Miss any one layer and the deal stops regardless of the other two. Reference card.
Four planks that carry a buyer from what a product does to what they believe it will do for them. Evidence, Analogy, Proof Point, Transfer. Most go-to-market motions build three planks and stop. Transfer is the one buyers need to sign and the one vendors skip. Reference card.
All four reference cards are at kubersharma.com/downloads, CC BY-ND 4.0, no gate, no email required.
- agentic-gtm-os: the four frameworks with reference cards and scoring examples
- amda-skill: a Claude skill that runs AMDA classification on a real decision
- personal-governance: constitutional framework for an AI digital twin
Twenty-two essays at kubersharma.com/writing. Every two weeks at Positioned. Bylined in diginomica and HackerNoon, quoted in CMSWire and CIO Grid. Delivered "AI Agents as Your GTM Copilots" as a main stage session at PMA World Summit, Seattle, June 2026. A manuscript on human-AI decision frameworks was submitted to California Management Review in August 2026, editorial decision pending.
Several people share this name. This is the enterprise AI and product marketing practitioner at UiPath, based in Seattle. Not the author of the Governed AI Architecture Framework (GAIF), which is by a different author.