Before the strategic partnership between UJJI AI and Google Cloud, the development of a single corporate training, especially for banks and financial entities, struggled with slow and linear production cycles, requiring a few weeks on average. This process relied on a chain of content creation specialists, approvals, and launch steps which created a “funnel”, limiting distribution times in organizations operating in a highly volatile industry.

The challenge for UJJI AI was to integrate innovative technology into its platforms, preserving its competitive edge, and always scaling content production with a secure and governed infrastructure. The rapid growth of the company’s client base and the increase in data security and privacy requirements evidenced that the legacy architecture was no longer able to support automation and scale demands.

The new solution meant a huge leap in terms of speed and computing power to handle dense multimodal data, such as long four-hour videos and complex technical documents. This ecosystem required seamless, latency-free, rigorous, and accurate execution to offer a comprehensive service experience.

The company had to create an infrastructure that removed the technical barriers between knowledge sources and employees needing to be trained. With this in mind, Google Cloud offered UJJI AI not only a technological platform, but also an engineering partnership to integrate Gen AI models directly into the company’s workflows, turning technology into an invisible enabler instead of an obstacle.

Adopting Gemini Enterprise Agent Platform allowed UJJI AI to shift the paradigm around customer productivity. Responsiveness in the creation of legal and administrative materials was significantly improved: repetitive, slow tasks, such as preparing trial summaries and meeting minutes, which used to take 40 minutes, now take less than a minute with the help of Gemini.

This agility boost breaks the market's standardized, traditional model, where generating one hour of training material requires, on average, weeks of specialized work. Automating the organization and summary creation around raw materials with Liz enabled UJJI AI to accelerate time-to-launch for new learning paths from two weeks to less than one hour. This also resulted in an 85% reduction in production costs.

Streamlining these processes freed T&D teams and managers to focus on what really matters: human development strategies and decision-making. Apart from the operational efficiency gains and cost savings, totaling about BRL 400,000 over three years, the project consolidated an innovation culture where technology is seen as an ethical and secure ally in one of the largest banks in Latin America.

The possibility of processing complex, multimodal data in a few seconds guaranteed quick, problem-solving actions, even in critical and high-pressure moments, like shift rotations and audits.