🥹 Thanks Anik for this thorough comment. I didn't even start advertising this blog, so receiving an input like yours gives me a lot of confidence.
You're right! Deployment automation with CDK can be quite interresting, especially for teams.
Adding monitoring & alerting is indeed a must-have. Currently, the cloudformation template I provided only logs the lambda's invocation lifecycle (due to SAM globals: "Tracing: Active" and "LoggingConfig.LogFormat: JSON")
- Alerting on Cloudwatch must be added
- If using cloudflare workers and Grafana Cloud, alerting in Grafana Cloud must be added using Alertmanager.
Both should be quite doable using AWS & Grafana Cloud dashboard, I'll check that
This is a fantastic, thorough walkthrough for building a scalable and cost-effective email sending system using serverless architecture and third-party APIs. I really appreciate the practical focus on handling rate limits via queues, which is often overlooked but crucial for reliability at scale.
The comparison of email APIs is very helpful, especially highlighting Resend’s generous free tier and dashboard features. Also, integrating AWS Lambda with SQS for event-driven processing is a great pattern to avoid direct API rate limit issues and to build a fault-tolerant system.
I like how the article balances practical implementation details (with schema validation and retry logic) and architectural insights (like microservices on Cloudflare Workers and queue decoupling).
Would love to see a follow-up covering deployment automation with SAM or CDK, and maybe tips on monitoring or logging email delivery success and failures.
Overall, a must-read for anyone wanting to send transactional emails reliably without overcomplicating the stack or cost. Thanks for sharing!