
Agentic AI in the Back Office: Why AI Value Starts With Your Processes | Flex x Salesforce
Enterprise AI has hit an inflection point, and it all starts with your processes, not your models. In this episode of The Supply Chain Revolution, created in paid partnership with Salesforce (#ad #SalesforcePartner), Sheri Hinish, the Supply Chain Queen, sits down with John Lane, Vice President of IT Application Solutions and Implementations at Flex, an Agentforce Operations customer. John shares how the global manufacturer is using Salesforce's Agentforce Operations, powered by Regrello, to run processes with AI from the ground up across its core back-office operations. Flex runs a large global manufacturing footprint supported by an equally large back office, where repetitive work such as financial transactions, engineering change orders, and approvals must run the same way everywhere. John explains why Flex standardized its end-to-end processes before adding agents, because putting AI on a poor process only gets you to the wrong place faster. He describes how his team now builds agentic workflows in minutes to hours instead of weeks to months, how the Systems Integration Agent connects to legacy applications without a costly platform migration, and why that approach ends the integration tax that has slowed enterprise IT for decades. The results are concrete. Flex is automating 95% of its work order approvals and roughly 70% of its third-party risk assessment activities. John also explains where people stay firmly in the loop, from customer-facing variance decisions to final security sign-off, and why faster delivery now lets his team pursue smaller use cases that were never worth the effort before. Key takeaways: • AI value depends on process discipline, not model horsepower. • Flex sequences process engineering, digitization, and agents so that solutions scale beyond isolated pockets. • Agentforce Operations orchestrates approvals, document analysis, and handoffs inside a single workflow. • The Systems Integration Agent reads legacy application screens and produces a reusable MCP layer for future integrations. • Human judgment remains essential for claimable work order variances and for final vendor security decisions. • Clean, accessible data is the other half of turning AI hype into measurable business value. Chapters: (00:00) Enterprise AI's inflection point: why AI value starts with your processes (00:37) Introducing Flex and agentic AI in the back office (02:00) Why process mapping comes first in AI deployment (03:38) How Agentforce Operations streamlines back-office processes (05:54) Connecting legacy systems with AI: ending the integration tax (07:51) Human in the loop: oversight in automated processes (11:45) Redirecting saved time for greater impact (14:01) Turning AI hype into measurable business value (15:56) The back office is next Learn more about Agentforce Operations: https://www.salesforce.com/agentforce/operations/ Connect with John Lane on LinkedIn: [John's LinkedIn URL] Learn more about Flex: https://flex.com Follow Sheri Hinish, the Supply Chain Queen: https://supplychainqueen.com Watch the full conversation on YouTube: https://www.youtube.com/@supplychainqueen If this conversation was valuable, follow The Supply Chain Revolution and share this episode with an operations leader who is ready to stop paying the integration tax. Topics: Agentforce Operations, Salesforce, Regrello, agentic AI, AI agents, agentic enterprise, enterprise AI, process automation, back-office automation, supply chain operations, legacy system integration, human in the loop, AI governance and trust, manufacturing, Flex.







