Everything in Backend Dev, page 35
Python's async/await: Concurrent, Not Parallel
async/await lets a single Python thread juggle multiple tasks, pausing one to work on another while it waits for I/O. It's ideal for network requests or database queries. The footgun: it won't speed up CPU-bound tasks, it only helps with waiting.
Python Coroutines: Functions You Can Pause and Resume
A Python coroutine is a function that can be paused and resumed. It yields control during I/O waits, allowing other tasks to run instead of blocking the program. The main footgun: calling an async function does nothing; you must await it to run it.
The `with` Statement: Python's Automatic Cleanup Crew
A context manager is Python's automatic cleanup crew. It uses the with statement to guarantee setup and teardown code runs, even if errors occur. It's essential for files and database connections.
Python's `yield`: Functions That Pause and Resume
Python's yield creates a generator: a pausable function that produces values on-demand, saving memory. Use it for large files or infinite sequences. The footgun: a generator is a one-time-use iterator; you can't loop over it twice.
Python Decorators: Functions that Wrap Functions
A decorator is a function that wraps another function, adding behavior without modifying the original code. They're used for caching, logging, or access control. The main footgun is forgetting that decorators run at definition time, not call time.
Python Enums: Give Names to Magic Numbers
Python's Enum gives meaningful names to "magic numbers" or strings. Use it for fixed sets of options like statuses or categories to make code self-documenting. The footgun: don't compare members to raw values; compare member to member for type safety.
Python Packages: Grouping Modules with __init__.py
A package groups Python modules under an import name. A regular package has an __init__.py file, which runs when it is first imported. Namespace packages can omit this file. Use packages to organize related code.
Python Data Classes: Write Less Boilerplate
Python's @dataclass decorator writes boilerplate code like __init__ and __repr__ for you, turning a class with type hints into a data container. Use it for API payloads or simple records.
Python Type Hints: Documentation Your Linter Can Read
Type hints describe expected values, but Python itself does not enforce them while running. Static checkers and IDEs use the annotations before execution; libraries such as FastAPI can also inspect them for validation and API documentation.
Production Secret Management: Inject, Don't Store
Treat secrets like temporary credentials, injected at runtime, not stored with your code. This applies to database passwords and API keys in production. The biggest footgun is using .env files; they are a dev convenience, not a security model.
Heap Snapshots: Finding Node.js Memory Leaks
A heap snapshot is a photograph of your app's memory. Use it to diagnose leaks by comparing snapshots over time to see which objects grow. The big footgun: taking one freezes your app and can double memory usage, risking a crash in production.
Sinon.JS: Isolate and Inspect Code for Unit Tests
Sinon.JS lets you replace real functions with test doubles to check *if* and *how* they were called. Use it to fake network requests or control timers. The biggest footgun is forgetting to restore fakes, which causes tests to leak state and fail unpredictably.
JWTs for Stateless API Authentication
JWTs enable stateless authentication: your server verifies users via a self-contained, signed token instead of a session store. This is ideal for distributed APIs. The biggest footgun is storing refresh tokens in localStorage; use HttpOnly cookies instead.
Passport.js: The Gatekeeper for Your Routes
Passport.js is a gatekeeper for your Node.js routes, authenticating requests before your application logic runs. It uses pluggable "strategies" for different login types, like local passwords or Google OAuth. The footgun is misconfiguring failure handling.
Cookie-Based Sessions: Server-Side State, Client-Side ID
Think of a session cookie as a coat check ticket, not the coat itself. The server stores your data and gives you a unique ID to carry in a cookie. This is how Express.js tracks user state across requests.
Never Trust Client Input: API Validation
Think of API validation as a bouncer for your server, checking every incoming request's ID before it can access your application logic. Use it in any Express route that accepts user input to prevent bad data from hitting your database or causing errors.
package-lock.json: Your Dependency Blueprint
package-lock.json is a blueprint for your node_modules, ensuring everyone on your team installs the exact same dependency versions. It's auto-generated by npm to prevent 'works on my machine' bugs. The footgun is ignoring it or manually editing it.
Non-Blocking I/O: Don't Block the Event Loop
Non-blocking I/O lets your program do other work while waiting for slow operations like network requests. It's the core of Node.js, allowing a single thread to serve many users.
CI/CD Pipelines for Node.js Applications
A CI/CD pipeline is an automated assembly line for Node.js code, installing dependencies, running tests, and packaging your app for deployment. This is standard for any professional project, but a common footgun is not caching dependencies, leading to slow…
Docker Compose for Multi-Container Apps
Docker Compose is a conductor for your containers. Instead of running each service manually, you define your app and its database in one YAML file and launch them together. This is standard for local Node.js/Postgres development.
We are hiring for this. Every open role lists the topics its interview covers, so you can prepare for the real thing rather than guessing.
See open roles