Lambda Fundamentals

Prerequisites: an AWS account with an IAM user or role that has programmatic access; AWS CLI v2 installed and configured (aws configure); basic comfort with either Python or Node.js.

Version note: verified against docs.aws.amazon.com/lambda, July 2026. This project targets python3.13 and nodejs22.x — both current, fully-supported managed runtimes. Avoid python3.9/3.10/3.11 and nodejs18.x/nodejs20.x in new work: 3.9 stopped receiving patches Dec 2025, nodejs20.x stopped receiving patches April 30 2026 (function creation blocked from Aug 31 2026, updates blocked from Sep 30 2026), and 3.10/3.11 are on a similar glide path. Run aws lambda list-runtimes or check the Lambda console's runtime dropdown for the definitive current list before you build anything beyond this tutorial.

A Lambda function is a single, stateless handler function that AWS runs for you on demand, inside a managed "execution environment" (a lightweight Firecracker microVM) that AWS provisions, patches, and tears down without you ever touching a server. You write a handler with a fixed signature — (event, context) — package it up, and Lambda invokes it whenever a configured trigger fires: an HTTP request, a new message on a queue, a file landing in S3, a schedule, or a direct API call.

The first invocation on a fresh execution environment pays a startup cost called a cold start (you'll dig into that in depth in the Cold Start Optimization project); subsequent invocations on the same warm environment skip that cost. Understanding that distinction — and the fact that Lambda, not you, decides when to reuse or discard an environment — explains most of what feels surprising about Lambda at first (global variables persisting between invocations, database connections needing to be re-validated, etc.).