{"slug":"tbc-bank-operationalizes-trusted-data-with-lakebase","url":"https://findausecase.com/use-cases/tbc-bank-operationalizes-trusted-data-with-lakebase","title":"TBC Bank Operationalizes Trusted Data with Lakebase","description":"TBC Bank, the largest banking group in the Caucasus region, built a Lakehouse on Databricks and adopted Lakebase and Databricks Apps to move from on-premises SQL Server instances and month-long reporting cycles to self-service analytics and AI-driven applications, including a web-based AI chatbot and AutoML-based credit risk scoring. Credit risk model deployment fell from 14 weeks to two days, and more than 600 users regularly query governed data through Genie.","company":"TBC Bank","industry":"Financial Services","country":"Georgia","aiCapabilities":["Machine Learning","Conversational AI"],"technology":["Databricks Platform","Lakebase","Databricks Apps","Unity Catalog","Genie","Databricks SQL","AutoML"],"deployment":"Public Cloud","problemStatement":"TBC Bank relied on on-premises SQL Server instances with month-long reporting cycles and no platform for advanced analytics. A single analysis could take up to a month across disconnected datasets, credit risk scoring models took 14 weeks to deploy — slowing loan disbursement in a market where speed to decision directly affects profitability — and the bank had no way to serve transactional workloads without spinning up standalone databases, duplicating security controls and increasing operational overhead.","solutionApproach":"TBC Bank built a Lakehouse on the Databricks Platform, migrated all workloads to the cloud and adopted Unity Catalog as its governance layer, then added Lakebase as an operational database layer bridging governed data with applications needing fast reads and writes. Using Databricks Apps, the team delivered a reverse-ETL pipeline and a web-based AI chatbot with Lakebase as the transactional backend, adopted AutoML for credit risk scoring, and rolled out Genie for self-service conversational analytics over governed data.","businessValue":"Credit risk model deployment fell from 14 weeks to two days including model risk validation, analyses that once took a month now take minutes through Genie, more than 600 users regularly query governed data through Genie (roughly 20% of headquarters staff today, with adoption expected to reach 50% of headquarters staff by year-end), the data analyst team was reduced by 50% with time reallocated to higher-value work, and the four-person DevOps team deployed Lakebase in a couple of hours.","evidence":{"band":"high"},"sourceUrl":"https://www.databricks.com/customers/tbcbank/lakebase","dates":{"publishedAt":"2026-08-17T06:48:35.535Z","publishedAtSource":"ledger","updatedAt":"2026-08-25T21:05:43.193Z"},"license":"Open for reading and citing with a link to https://findausecase.com/use-cases/tbc-bank-operationalizes-trusted-data-with-lakebase. Bulk republication requires permission."}