What AWS services have you worked with? What are the limitations of AWS Lambda? What limitations do you see in Amazon API Gateway?

I have worked with several core AWS services, including Lambda, API Gateway, S3, DynamoDB, CloudWatch, and Glue. Lambda is a serverless compute service that runs code in response to events. Its main limitations are a 15‑minute maximum execution time, a 10 GB deployment package size (including layers), memory limits up to 10 GB, and a default concurrency limit of 1,000 (which can be increased via request). Cold starts can add latency, especially for languages with large runtimes. Lambda also has limits on the number of concurrent connections to external services and on the size of environment variables. Amazon API Gateway, which exposes Lambda functions or HTTP endpoints, has its own constraints: the maximum request body size is 10 MB, the maximum integration payload is 10 MB, and there are per‑account throttling limits (default 10 000 requests per second). It also imposes a 29‑second maximum integration timeout for Lambda proxies. These limits can affect high‑throughput or large‑payload applications, so careful design (e.g., using S3 for large uploads or using WebSocket APIs for real‑time traffic) is required.