How do you invoke a Lambda function manually?

You just deployed a Lambda function and want to test it quickly in the AWS Console — walk me through the steps you'd take and what you'd look for in the result.

A teammate asks you to invoke a Lambda from your laptop using the AWS CLI with a JSON payload — what command would you run, and how do you verify it executed successfully?

What happens if you try to invoke a Lambda function that doesn't exist, or one you don't have permission to invoke? What error would you see?

You're building a backend service that needs to trigger a Lambda asynchronously and continue without waiting — how would you invoke it, and how do you handle the fact that you won't get the function's return value immediately?

A manual CLI invocation of your Lambda is timing out after 30 seconds — what are the likely causes, and how would you debug whether it's a function issue, a payload issue, or a configuration issue?

You need to test a Lambda with a complex payload shape that includes nested objects and arrays — how do you pass that payload via CLI, and how do you confirm the function parsed it correctly?

Your team needs to manually invoke a Lambda 10,000 times for a one-time data backfill — how would you design the invocation pattern to avoid throttling, handle partial failures, and track progress?

You're building an internal CLI tool for developers to invoke Lambdas with different payloads and invocation types — what invocation modes (sync vs async) would you expose, and what guardrails would you add to prevent accidental production impact?

During incidents, engineers manually invoke a critical Lambda to remediate issues — how would you add audit logging, approval gates, and dry-run capability without slowing down emergency response?

Your organization is migrating from ad-hoc manual Lambda invocations via CLI to a centralized internal platform — what architectural decisions would you make to ensure safety, observability, cost attribution, and developer experience across teams?

Multiple teams need to invoke each other's Lambdas manually for integration testing — how would you design a self-service invocation framework with proper access controls, quota management, and audit trails?

You're evaluating whether to replace recurring manual Lambda invocations with EventBridge schedules or Step Functions — what factors would drive your decision, and how would you migrate existing workflows with zero downtime?