AWS Lambda Reviews & Product Details

It’s serverless, so we don’t have to worry about managing servers—AWS handles that for us. We use AWS Lamda for running our performance tests, we don't have to keep server running, whenever we need we can spin up and run the tests for API's, good for short running tasks. Easy to maintain and integrate it from command line. Price wise also we don't spend much since it's costing us for what we run. Later to trouble shoot issues, we cna go to UI and check the aws cloudwatch logs, AI can help as well if we want are trougble shooting something specific, and onboarding same process for other env is easy. Review collected by and hosted on G2.com.

I wish it could support more than 15 minutes; if they could extend it to 30 minutes, that would be great. Also, it takes a while to bootstrap and start running. Review collected by and hosted on G2.com.

What I like most about AWS Lambda is how easy it makes running applications without worrying about servers. It automatically scales based on demand, saves time on infrastructure management, and helps reduce costs because you only pay when your code runs. It integrates smoothly with other AWS services, making development much faster and simpler. Review collected by and hosted on G2.com.

One thing I dislike about AWS Lambda is that debugging and troubleshooting can sometimes be challenging, especially for complex applications. Cold starts may also cause slight delays for infrequently used functions, and managing execution time limits can be restrictive for long-running workloads. However, for most use cases, the benefits outweigh these limitations. Review collected by and hosted on G2.com.

I frequently use AWS Lambda in our project. We mainly use it for event-driven automation in our workflow. The best part is that once files arrive in S3, we’ve set it up to trigger automatically. This then automates the validation process, which I like the most. On top of that, its integration with other AWS tools is fast. Review collected by and hosted on G2.com.

I don’t have any major dislikes about AWS Lambda. Overall, it’s a good product and has worked well for me. Review collected by and hosted on G2.com.

It is easy to configure. It has lots of user controls like run time, timeout, source event, layers, and it is pay-as-you-go. It gives immense power for fan-out scaling. Also, it is compatible with lots of AWS native services and integrates well with them, like API Gateway, S3, SNS, SQS, CloudWatch, and many more Review collected by and hosted on G2.com.

Lambda has an upper limit on runtime, so you will have to design your system accordingly. Also, debugging becomes quite messy because there will be many invocations happening simultaneously. Review collected by and hosted on G2.com.

It is a very well supported tool that provides great infrastructure for running background jobs and tasks periodically. It has great support with other AWS services and is easy to run with a lot of great documentation. Review collected by and hosted on G2.com.

I don’t like that the maximum function time is limited to 15 minutes. I really hope this can be increased, because it doesn’t always accommodate bigger jobs. Review collected by and hosted on G2.com.

The native, abstract integration with the AWS event ecosystem.

Lambda feels like the ultimate connective tissue. Rather than writing boilerplate polling logic or managing persistent connections, you get clean, out-of-the-box event mappings for S3 hooks, DynamoDB Streams, SQS queues, and Kinesis. Being able to write a tightly focused handler function and then let the platform handle upstream ingestion, batch sizing, and retry behavior out of the box completely changes how quickly you can ship decoupled microservices. Review collected by and hosted on G2.com.

The most frustrating part of working with AWS Lambda, for me, is dealing with its architectural guardrails—especially cold starts and the strict 15-minute execution limit. For synchronous, user-facing APIs, a cold start can cause a noticeable latency spike, and it’s even worse when the function runs inside a VPC. That reality forces careful mitigation (like Provisioned Concurrency), but doing so undercuts the “scale-to-zero” cost benefit.

On top of that, the hard timeout means Lambda simply isn’t a fit for long-running batch jobs or continuous processing. You end up having to split work into more complex, decoupled Step Functions workflows or offload it to ECS/Fargate, which adds architectural complexity that feels unnecessary. Review collected by and hosted on G2.com.

No infrastructure management: You just deploy your function and AWS handles servers, scaling, and maintenance.

Automatic scaling: It scales instantly from zero to thousands of requests without configuration.

Cost efficiency: You only pay for execution time, not idle server time. Review collected by and hosted on G2.com.

Cold starts: Functions can have latency delays when they haven’t been used recently, which affects performance in time-sensitive apps.

Execution limits: There are strict timeouts (max runtime), memory constraints, and ephemeral storage limits, which make it unsuitable for long-running or heavy compute tasks. Review collected by and hosted on G2.com.

What I like most about AWS Lambda is that it allows you to execute business logic and automations without having to manage servers. It is very useful for building lightweight integrations, responding to events, and decoupling processes within an AWS architecture. I also appreciate that it integrates well with other services like S3, EventBridge, Step Functions, IAM, and CloudWatch, which greatly facilitates the construction of serverless workflows. For specific tasks, event processing, and auxiliary components of pipelines, it is very practical and scalable. Review collected by and hosted on G2.com.

What I like least is that debugging and troubleshooting can become more complex when the function depends on multiple services, permissions, or external configurations. There are also limitations that require careful design of certain use cases, such as maximum execution times, handling heavier packages, or specific dependencies. In some scenarios, monitoring errors or understanding integration failures takes longer than expected. Review collected by and hosted on G2.com.

You pay what you use, For projects with spiky traffic—like an app that only gets hits during business hours or special sale like Black Friday, this saves us a fortune. Review collected by and hosted on G2.com.

If you've ever tried to trace a single request across five different Lambda functions using CloudWatch's default UI, you know it's a nightmare. To get decent observability, you usually have to pay for a third-party tool like Datadog or New Relic, which adds another layer of cost.

Second, it should only cost for execution and not for cold start when code is actually waking up Review collected by and hosted on G2.com.

The serverless model — no infrastructure to manage, automatic scaling, and you only pay for what you use. It makes spinning up lightweight microservices and event-driven workflows very straightforward Review collected by and hosted on G2.com.

Cold starts can be frustrating for latency-sensitive workloads. The 15-minute execution limit is restrictive for longer-running tasks, and debugging in production is harder compared to traditional server-based setups. Review collected by and hosted on G2.com.