What Happens to Tribal Knowledge When Dispatchers Retire?

For decades, the freight industry has heavily invested in technology for large freight companies, leading to the adoption of sophisticated transportation management systems and software integrations. However, many small-to-medium-sized businesses still rely on outdated TMS platforms, often paying enterprise prices for software that may not cater to their specific needs.

This technology imbalance is the starting point for Evolutionary Operating System, Inc. (Evos), an autonomous AI operations company launching on October 14. CEO Urav Shah explains that decades of technology investment has mostly benefited the enterprise end of the market, leaving small and mid-size logistics companies with aging systems that are not tailored to their unique operations.

These systems can handle routine tasks, but fail to address edge cases, such as judgment calls that do not fit standard workflows, which are typically handled by people. Shah argues that this creates a structural capacity problem, especially as the industry's most experienced operators retire, taking their tacit knowledge with them.

To tackle this issue, Evos offers a platform that configures itself around a given business, learning the judgment used by its team to work through edge cases. The platform then handles this work on behalf of the company, rather than just flagging it for a person to complete. This execution layer allows Evos to run entire desks, such as dispatch, track-and-trace, or billing, end to end, on top of an organization's existing systems.

Shah emphasizes that the challenge lies not only in building software that can act but also in building software that acts like a specific company. Every logistics operation runs differently, and these variations appear in the edge cases that consume an ops team's day. Evos uses large action models, trained on patterns of decisions and behavior rather than text, ensuring that each customer has its own model trained solely on that company's historical data.

This approach preserves the company's "tribal knowledge" and avoids potential IP infringements. Furthermore, Evos aims to address the industry's aging workforce by preserving the knowledge of retiring dispatchers and fleet managers before it is lost. The platform can go live in under 24 hours from the first conversation, using a library of over 150 pre-built integrations with common freight systems, allowing for rapid deployment without migrating existing data.