Plata’s fintech data warehouse fuels rapid growth in Mexico
Ivan Shchukin joined Plata three and a half years ago, when the Mexican fintech had no data warehouse. Today, it has a department of more than 50 people, near-real-time data processing, and an ambition to render traditional business intelligence software obsolete. This shift represents a strategic wager on data as a weapon—one that helped Plata secure its banking license and expand from credit cards and cashback programs into small-business lending and investment products.
Shchukin’s unusual background, combining electrical engineering, project management, and data analysis, positioned him uniquely for the challenge. His CTO, Andrey Shelekhin, called him the right person to construct something rare in fintech: a data warehouse that balanced completeness with speed. The system now tracks every modification to every record, preserving full version history, while most competitors rely on 24-hour refresh cycles. “We want to collect all data in the company with full versioning,” Shchukin explains. “Everyone wants this, but most players in the market are afraid of it. I had no limitations and no fears.”
The one-hour latency goal went beyond technical requirements; it was a business imperative. In retail banking and fintech, the ability to identify and respond to problems, such as a spike in failed transactions or a dip in customer engagement, within hours rather than days creates a measurable competitive advantage. “We can measure our business very precisely, like checking a person’s temperature,” Shchukin says. “It’s the main reason we’ve been so successful in Mexico; we know what is working and what isn’t, and we can react in hours, not days.”
Today, the warehouse underpins approximately 80-90% of the company’s data processes, all running with automated quality checks and alerts. This efficiency has enabled the company to maintain a lean headquarters despite rapid expansion. The team has grown from eight members upon Shchukin’s arrival to more than 50 people organized across roughly 10 specialist teams.
AI tools replace rigid BI dashboards
Plata’s approach diverges sharply from industry norms. While most fintechs construct data warehouses around traditional dashboards, the company is moving away from this model. The emergence of large language models has led Shchukin to reconsider how analysts interact with data entirely. Rather than building fixed dashboards in dedicated BI platforms, analysts would instead use AI tools to generate bespoke visualizations on demand, written in a Python framework called Streamlit. “We think that within the next year, every analyst will write their own small BI tools using AI agents,” Shchukin says. “It will be a total disruption, a completely different experience of working with data.”
The transition carries risks, however. AI-generated insights demand meticulous metadata, structured descriptions of datasets and their relationships.
Related Post: Bank of America launches seamless cross-border payments for businesses
This shift mirrors broader trends in enterprise data management, where rigid business intelligence tools are being replaced by flexible, AI-assisted workflows. Currently, Plata’s staff handle multiple tools to perform data-related tasks. Consolidating these into one AI-assisted environment is among Shchukin’s primary objectives for the year ahead. “We want just one window for any task,” he explains.
Even internal support is evolving. While Plata has not shared specific automation targets, the company has deployed an AI agent to resolve routine inquiries by scanning a knowledge base.
Latin America expansion hinges on data speed
Looking ahead, Shchukin’s priorities include expanding the warehouse into additional Latin American markets and ensuring full AI tool adoption across his team. Currently, about 60% of the team incorporates AI into daily workflows.
The next phase of Plata’s data strategy will determine whether its emphasis on speed, automation, and AI-driven analytics can extend beyond Mexico. The company’s banking license and regional growth plans depend on proving that a data warehouse designed for near-instant insights, operated by analytics engineers rather than traditional data scientists, can surpass legacy systems in performance.
Plata’s data warehouse team now numbers over 50 specialists, organized into roughly 10 teams. This expansion reflects a deliberate shift from traditional data science toward analytics engineering, a discipline that bridges data infrastructure and business analysis. Unlike data scientists, who often concentrate on modeling and statistical work, Plata’s analytics engineers design and maintain pipelines that convert raw data into actionable insights. Their work ensures datasets are structured, versioned, and ready for immediate use by analysts, executives, or AI systems.
