Credit Risk Scorer Solution

In the competitive financial landscape, the ability to quickly and accurately assess credit risk is crucial for financial institutions and banks. The Credit Risk Scorer Solution addresses this need by automating the credit rating assessment process. By uploading relevant documents such as credit proposals, project plans, and annual reports, the solution extracts and normalizes entities from these inputs. It retrieves core details, including financial data and credit history, from internal banking systems and databases. Additionally, it uses tools like Tavily search to gather up-to-date, external information from open-source web. The processed inputs are then analyzed by an assembler agent that applies business rules to recommend a credit decision with confidence scores and rationale. This automation enhances efficiency, reduces manual data entry and analysis errors, and improves turnaround times for credit assessments and decisions. The AWS services which were used are Amazon Bedrock, AWS Lambda, Amazon DynamoDB, Amazon S3, Amazon EC2 or AWS Fargate, AWS Kendra and Apprunner.