Kita

Kita addresses the challenges of open finance in emerging markets like the Philippines, where traditional banking infrastructure is limited. A significant portion of the population lacks access to banking services, and crucial financial data resides in unstructured documents such as e-wallet records, bank statements, and utility bills. This unstructured data necessitates manual review by credit and risk teams, resulting in delayed decision-making, increased operational costs, and restricted lending volumes. Existing Optical Character Recognition (OCR) solutions often fail when processing noisy, real-world documents and still require human verification.

Kita offers a document intelligence platform specifically designed for lending in these markets. The platform focuses on extracting key signals that drive lending outcomes in emerging and underserved domestic markets, including the Philippines, Indonesia, and Mexico. By employing a layered system that leverages vision-language models and computer vision, Kita surpasses traditional OCR methods. It transforms complex borrower documents into fraud-checked, decision-ready signals that lenders can directly utilize in their underwriting processes.

Kita operates as a learning system, connecting document-level signals with repayment outcomes. This allows the models to continuously enhance fraud detection and risk assessment capabilities over time. As lenders make distinct underwriting decisions, the system learns and improves, creating a compounding advantage. The company secured a six-figure paid pilot with a Philippine lender shortly after building its prototype and has since expanded its customer base across multiple Southeast Asian markets, consistently delivering new features and achieving substantial month-over-month growth.