Ramp engineering leader Arman Vaziri describes go-to-market orchestrationAI go-to-market orchestration coordinates customer data, decisions, automated actions, human tasks, and communication channels around a commercial outcome. as the ability to express a campaign or experiment as intentAn intent-based AI workflow starts from a desired outcome and determines the coordinated tasks needed to reach it., then distribute its execution across outbound, advertising, web and product channels. The goal is to reduce the operational coordination that slows otherwise useful ideas.
Ramp built an internal customer data platformA customer data platform unifies customer information from multiple sources into profiles and datasets that other systems can use. that combines CRM, product, enrichment, web, interaction and buying-signal data. Postgres preserves transactional relationships, warehouse jobs add offline computation, and embedded unstructured information lets agents retrieve only the account context they need.
Vaziri explains how durable Temporal workflowsDurable workflow execution records an AI workflow's progress so it can recover and continue after interruptions instead of starting again., scoped tools, human review and reusable skills support specific tasks such as meeting briefs and CRM updates. Ramp then exposes the same foundations through an internal MCP layerModel Context Protocol is a standard way for AI applications to connect with external tools and data sources through consistent interfaces., allowing employee experiments to reveal new workflows that the engineering team can productionize and distribute more broadly.
Watch on YouTube
