Automated Account Monitoring Pipeline at Scale
Playwright · ECS Fargate · SQS · Lambda · S3 · EventBridge · CDK
Built a fully automated scraping and monitoring pipeline that manages N independent Securus accounts, records video sessions, processes them through an SQS queue, and stores results in S3 — all orchestrated serverlessly on AWS.
What needed solving
The client needed to monitor activity across dozens of Securus accounts simultaneously, capture video recordings, and process them without manual intervention. Manual workflows were slow, error-prone, and couldn't scale.
How we built it
We designed a serverless-first architecture on AWS. A Playwright scraper runs in ECS Fargate containers, one per Securus account. An EventBridge Scheduler triggers Lambda to kick off scraping jobs on a schedule. Results flow through SQS — a producer Lambda enqueues jobs, a consumer Lambda processes them. Video recordings are stored in S3. A separate Update-State Lambda keeps account state in sync. Everything is deployed via AWS CDK.
Layer by layer
System Overview
What we delivered
- 100% automated — zero manual monitoring required
- Scales to N accounts with no architecture changes
- Video recordings reliably captured and stored in S3
- Full audit trail via CloudWatch and DynamoDB
- Deployed and reproducible via CDK in minutes
Tech Stack
Build something similar?
We can architect and build this kind of pipeline for your use case.
Book a Call →