Serverless event-driven architecture enabling engineering teams to process millions of daily events with near real-time visibility, strong resilience, and robust financial monitoring.

This project demonstrates a scalable, serverless, event-driven architecture on AWS, designed to process and monitor high volumes of invoice events (over 86 million daily) with near real-time observability, cross-Region controls, and automated alerting for stuck events. The solution is ideal for organizations seeking business-level insights and real-time visualization of event flows, as well as comprehensive historical retrieval for auditing and compliance.

For a deep-dive architectural walkthrough, see the AWS Architecture Blog: Implement event-driven invoice processing for resilient financial monitoring at scale and the GitHub repository.

Amazon API Gateway: Receives client events via REST API, supports throttling and error handling.

Amazon EventBridge: Custom event bus for intelligent routing and event archiving/replay.

Amazon SNS: Topic-based fanout for event distribution to downstream consumers.

Amazon SQS: Decouples event processing, provides dead-letter queues (DLQs) for failed messages.

AWS Lambda: Serverless compute for event processing, scales automatically with demand.

Amazon Timestream: Time series database for real-time and historical analytics.

Amazon QuickSight: Dashboards and analytics for business and operations users.

Event Producers: API Gateway receives events from clients.

Event Routing: EventBridge routes events to SNS topics for fanout.

Event Consumers: SQS queues buffer events for Lambda consumers.

Business Intelligence: Timestream and QuickSight provide dashboards and analytics.

Cellular Architecture: Independent, self-contained deployment units (cells) for scaling and fault isolation.

Serverless: Managed services minimize operational overhead and scale automatically.

Highly Available: Multi-AZ resilience, automatic failover, and disaster recovery.

EventBridge: Custom event bus with rules for content-based routing and archiving.

SNS Topics: Partitioned by event type (e.g., invoice-ingestion, invoice-reconciliation, etc.), with subscription filters for granular control.

API Design: RESTful endpoints for each invoice stage (/invoices/ingestion, /reconciliation, etc.).

Security: API keys, usage plans, AWS WAF, and throttling for protection and rate limiting.

Monitoring: CloudWatch Logs, AWS X-Ray, and custom metrics.

SQS Queues: FIFO queues per partition, each with a DLQ for failed messages.

Lambda Functions: One per queue, handling business logic, transformations, and Timestream ingestion.

Error Handling: Custom retry logic, exponential backoff, and alerting for stuck events.

Timestream: Multi-measure records for efficient time series analytics.

QuickSight: Dashboards for executives, operations, and finance, with ML-powered anomaly detection.

CloudWatch: Alarms for latency, error rates, and queue depth.

SNS: Alerting for different severities and automated escalation.

IAM: Least privilege roles for all services.

Encryption: AWS KMS for data at rest and in transit.

Compliance: AWS Config and CloudTrail for auditing and policy enforcement.

AWS account and credentials configured (aws configure)

AWS CDK installed

Node.js and npm

Before deploying, set your AWS Account ID as an environment variable:

Bootstrap the CDK in every deployment region:

npm run build – Compile TypeScript to JavaScript

npm run watch – Watch for changes and compile

npm run test – Run Jest unit tests

npx cdk deploy --all – Deploy the stack to your AWS account/region

npx cdk diff – Compare deployed stack with current state

npx cdk synth – Emit the synthesized CloudFormation template

See CONTRIBUTING.md for guidelines.

See CONTRIBUTING.md for security information.

This library is licensed under the MIT-0 License. See the LICENSE file for details.

Developed by Grey Newell, Senior Solutions Architect at AWS.

Connect with Grey on GitHub, X, or LinkedIn.