AWS also reduced decision time from three days to less than an hour and saved 100% in maintenance investments, in addition to minimizing infrastructure and development costs.
The financial technology provider Base39 revolutionized its operation through serverless architecture solutions AWS, a cloud computing and artificial intelligence services platform. By integrating generative artificial intelligence solutions, the fintech reduced loan analysis costs by 96% and decision time from three days to less than an hour. The model also reduced infrastructure costs by 84%, development costs by 75% and maintenance costs by 100%.
Prior to the migration, Base39 customers were performing loan assessments manually, with a maximum limit of 50 per person per day. This process resulted in high costs for the company and delays of up to three days in loan approval, compromising the customer experience. With this challenge in mind, AWS Consulting and the MongoDB for Startups program jointly proposed to Base39 an integration of Amazon Bedrock, a solution that allows developers to build and scale generative AI applications, with MongoDB Atlas Vector Search and AWS Anthropic Claude, which assists in data search and creates custom languages, implementing technologies capable of significantly improving efficiency, customer satisfaction, and innovation. To enable its generative AI solutions to deliver relevant and reliable results, Base39 used MongoDB Atlas Vector Search to implement the fine-tuning of retrieval augmented generation (RAG) for LLM with its proprietary data stored in MongoDB. With the integration, Base39 was able to improve financial solutions in Latin America.
Base39’s commitment to customer loyalty inspired the search for more flexible processes and time optimization, resulting in improved scalability and customer satisfaction. Because MongoDB Atlas uses a flexible and scalable document-based data model that supports virtually any data type and offers native vector capabilities, it was clear that it was the ideal database solution. “In just two weeks, AWS-based generative AI, powered by Amazon Bedrock and MongoDB Atlas Vector Search, transformed Base39 by significantly reducing costs and accelerating loan processing,” said Bruno Nunes, CEO of Base39. The results were achieved through Gen AI technology and the AWS team, which was available to continuously guide the use of the tools.
This transition to a serverless architecture has allowed Base39 to expand access to a range of AWS tools and services and eliminate additional layers of compute, improving performance and cost efficiency, while also reducing latency through customized solutions that offer ease of use. “Our goal is to enable technology integration and infrastructure support while our customers focus on delivering value as an organization to their consumers,” said Karina Lima, Head of Startups at AWS.
Another essential aspect of technology optimization solutions is security: with cloud-based backup, archiving, and document storage, large volumes of data can be managed efficiently and quickly for AI-driven analysis. To achieve this, Base39 used a combination of Anthropic’s Claude 3.5 Sonnet and Claude 3 Haiku on Amazon Bedrock, supported by LangGraph, an integrated development environment specialized in building AI agents. This combination allows Base39 to manage the complexity of credit analysis in a modular and adaptable way, enhancing its ability to deliver accurate and personalized credit solutions.
About Base39
Brazil-based Base39 provides technology solutions for financial services institutions (FSIs), including a loan management and loan analytics platform powered by generative AI and a fully serverless architecture using AWS Lambda and Node.js. Its unique ability to integrate with Brazilian payroll systems—often without APIs—simplifies payroll processing and enhances financial services. Its API catalog and white-label interface enable high customization and ease of use. Backed by Kinea Ventures, a subsidiary of Banco Itaú, Base39 leverages multiple AWS services and MongoDB Atlas for scalable, data-driven insights, improving loan approval accuracy and reducing costs. Its robust platform enhances loan management with streamlined processes and hyper-personalized customer experiences.
About Amazon Web Services
Since 2006, Amazon Web Services has been the world’s most comprehensive and broadly adopted cloud. AWS has continually expanded its services to support virtually any workload, and now has over 240 fully featured services for compute, storage, databases, networking, analytics, machine learning and artificial intelligence (AI), Internet of Things (IoT), mobile, security, hybrid, media, and application development, deployment, and management, across 108 Availability Zones within 34 geographic regions, with announced plans for 18 additional Availability Zones and six new AWS Regions in Mexico, New Zealand, Kingdom of Saudi Arabia, Taiwan, Thailand, and the AWS European Sovereign Cloud. Millions of customers—including the fastest-growing startups, largest enterprises, and leading government agencies—trust AWS to power their infrastructure, become more agile, and lower costs. To learn more about AWS, visit aws.amazon.com.
About Amazon
Amazon is guided by four principles: customer obsession rather than competitor focus, passion for invention, commitment to operational excellence, and long-term thinking. Amazon strives to be Earth’s most customer-centric company, Earth’s best employer, and Earth’s safest place to work. Customer reviews, 1-Click shopping, personalized recommendations, Prime, Fulfillment by Amazon, AWS, Kindle Direct Publishing, Kindle, Career Choice, Fire tablets, Fire TV, Amazon Echo, Alexa, Just Walk Out technology, Amazon Studios, and The Climate Pledge are some of the things Amazon has pioneered. For more information, visit amazon.com/about and follow AmazonNews.
EN 
PT 
