TRAXION

TRAXION stopped tying growth to headcount after Creai built MIND, a mobility system whose agents confirm trips and classify incidents. Trip confirmation went from 10% to 70%, planning that used to take weeks now takes a single click, and Creai deployed 27 sites with zero failures across 3 business units.

Before Creai

• Trip confirmations depended on enough people being available to handle them, tracked across phone calls and WhatsApp groups.
• Planning a route took weeks.
• Five separate systems (Bustrax, FICO, Siscot, SIGI, Billing) with no single source of truth.

With Creai

• 27 sites deployed, zero failures.
• Trip confirmation went from 10% to 70%.
• Planning now resolved in a single click.
• Running today across 3 business units, expanding to 24.

The case: TRAXION

TRAXION operates one of the largest mobility hubs in its region, coordinating a large, decentralized fleet of drivers and vehicles across constantly shifting routes and schedules.

It came in asking to modernize a system built on six-year-old technology, with a direct request: update it, make it faster.

There was also a reasonable amount of skepticism underneath: it wasn't the first time someone had promised that technology would fix something they had been trying to solve without success.

Industry context

In operations that run on volume (mobility, logistics, any business coordinating people and resources in real time) there's an almost automatic instinct when the workload grows: if trips, incidents, or coordination increase, the team needs to grow with them.

For a while, that logic works reasonably well. The problem shows up when that link between volume and headcount becomes structural, when the operation has no option but to grow in people because the system behind it offers no alternative.

Modernizing the technology without solving that dependency first produces something faster, but just as tied to the same hands to function.

Diagnosis: where the real problem was

Instead of running the plan as it arrived, Creai started with a full assessment of the operation before touching a single line of code.

The diagnosis showed that the real problem had less to do with how old the system was and more to do with how the team behind it actually worked. Every trip confirmation and every reported incident depended on enough people being available to handle it.

Much of that information lived scattered across phone calls and WhatsApp groups, with no central place to consolidate it.

The project traces back to an RFP TRAXION sent to several providers, asking for a transport management system built for moving people.

Creai was the only one that read it closely enough to catch the problem: it specified integrations with systems that no longer existed, and processes and roles that had already changed.

“I can't quote this development, because your document doesn't work.” — Rodrigo Velasco, project lead at Creai

That refusal to quote blind is what won Creai the consultancy to rewrite the RFP first.

Creai rewrote it before writing a single line of code, turning a 100-page document into a 250-page one with real user stories and hour estimates by role.

With that diagnosis, Creai identified 75 improvement opportunities within the operation, prioritized 11 high-impact ones, and deployed 7 specialized teams in parallel to address them.

That's where MIND came from: a mobility system with agents that confirm trips (Travel Companion), classify incidents (Incident Classifier), and coordinate tasks between them (Agent Orchestrator), along with predictive planning models that anticipate needs before they turn into emergencies.

MIND absorbed and replaced five systems that used to operate separately (Bustrax, FICO, Siscot, SIGI, and Billing), consolidating everything into a single source of truth.

The design started from a simple principle: Creai builds the technology around how the team works.

The agents take on the repetitive part of the operation that used to depend entirely on someone being available to do it by hand, and the people who coordinate it keep the decisions.

Implementation: 27 sites, zero failures

The real challenge was adoption: getting operators, coordinators, and supervisors across dozens of sites to drop Excel and the phone, and trust their day-to-day to a new system.

“It took us seven months to win the client over.” — Rodrigo Velasco

Creai defined adoption metrics before building, trained site leaders while the system was still being developed, rolled out in waves based on each site's risk level, and closed each site only with verified evidence that the operation was already working differently.

The result: 27 sites deployed, zero failures.

Conclusion

The TRAXION case shows that the link between growth and team size is a symptom of a system that offers no other option.

Before the implementation, more volume almost automatically meant more people. Afterward, the ability to grow stopped being tied to how many hands were available to sustain repetitive processes, because the system absorbed that part of the work.

The decision to diagnose before building is what made it possible to find the real problem, instead of automating an operation that would have kept depending on the same hands to function.

Treating adoption as part of the design, instead of a step after the technical implementation, explains why 27 sites closed without a single failure: the technology was ready before anyone was asked to trust it.

Where this transformation can go next

MIND already centralizes the data for the whole operation and already has agents in production confirming trips and classifying incidents.

That same foundation makes it possible to keep adding agents without rebuilding anything from zero: each new module inherits the Data Lake and the already-automated processes.

MIND's own roadmap already reflects this: growing on three fronts at once, new operational modules, more virtual agents, and additional predictive models, all on the same infrastructure.

The expansion from 3 to 24 business units is proof that the layered design works: each new business unit inherits a system that already knows how to operate, so the diagnosis and build work does not start from scratch.

That trust already produced a second deal, outside mobility entirely: Creai is now building 13 additional virtual agents on Silia for other divisions within TRAXION's parent group, a direct result of how the relationship played out on this first project.