CANCÚN, Q. ROO.- Artificial Intelligence (AI) only makes sense when it is used to solve real customer problems and not when it is applied automatically. This was the premise with which Andrés Arias, Senior Manager Customer Solutions at Amazon Web Services (AWS) , explained that the company’s strategy is built “from back to front”, placing the customer at the center and not the technology .
This vision shaped various mechanisms that allow for rapid experimentation, correction, and progress. During his participation in the 4th DCT Conference – Quálitas Infinite Possibilities , Arias explained that many decisions are considered reversible, known within the company as “two-way doors.”
Under this model, 100% information isn’t required to act; a reasonable level of certainty is sufficient to allow for cost-effective testing and adjustments.
In practice, this approach empowers teams to move with agility, which is essential in an environment like transportation, where fleet management requires immediate response to risks and operational changes .
Organizational culture was also a central focus of his presentation. Arias noted that at Amazon, small, autonomous teams, known as two pizza teams , are formed, which concentrate the ability to decide, execute, and measure results.
This scheme avoids bureaucracy and encourages speed of response , factors that are critical to transferring innovation into daily operations.
Regarding the relationship between innovation and failure, he acknowledged that invention cannot be separated from mistakes and cited the example of the Fire Phone, a failed project that generated millions in losses for the company in recent years, but whose lessons learned were leveraged in subsequent developments.
Beyond culture and processes, Arias explained that technological architecture also plays a decisive role. Amazon has transitioned over time toward a microservices model , which allows for the creation of fast, scalable, and low-cost solutions.
This structure makes it easier for teams to work in parallel, developing services that can connect to each other through interfaces, without relying on complex integrations.
For the transportation sector , this type of architecture opens the door to AI solutions that adapt to the needs of each fleet and can be turned on or off based on demand, reducing failure costs and accelerating adoption.
Artificial intelligence, viewed from this perspective, ceases to be an abstract promise and becomes a practical tool for simplifying processes , improving decision-making, and strengthening carriers’ responsiveness.
Comment and follow us on X: @karinaquintero / @GrupoT21
