Firebase
Welcome, champion! 🔥
You've mastered 0 of 78 lessons. Start your first lesson free — no signup needed.
About this course
Firebase lets a web app sign users in, store data, keep files and run backend code without managing a server, which is why so many first apps and side projects start on it. It is also easy to get wrong: data models that do not fit the queries, security rules left in test mode, and listeners that never unsubscribe. This online Firebase course on Abekus teaches the platform through one running example, a shared study-notes app called StudyCircle, in short reading-card lessons each followed by an MCQ quiz.
Who this Firebase course is for
It is written for front-end and full-stack developers, and for students building their first app with a hosted backend. You should be comfortable with modern JavaScript, including imports, functions, objects and async and await; the course does not teach JavaScript itself. By the end you can initialise Firebase with the modular SDK, sign users in and react to auth state, model, write, read and query Cloud Firestore data, write security rules that validate data and restrict access per user, upload files to Cloud Storage, add Cloud Functions and test everything locally in the Emulator Suite.
All examples use the Firebase JavaScript SDK with the modular API, where you import functions such as getFirestore and doc. The older namespaced API appears only in the lesson on migrating with the compat libraries. Security rules always start with rules_version 2. Code is shown as short text blocks rather than console screenshots, so nothing depends on where a button sits today. The course leaves out the native Android, iOS and Flutter SDKs, Analytics, Crashlytics, Remote Config, Messaging and the AI features, and it does not quote prices; plans are described only as Spark and Blaze. Firebase is a Google product, and this course is independent of Google.
How the lessons work
Each lesson is a short set of reading cards on one idea, usually with a few lines of code from StudyCircle, followed by a quiz of multiple-choice questions that often ask what a snippet will do. Lessons start with saving and reading a single note, then move through the modular SDK, Authentication, the Firestore data model, reads and writes, queries and indexes, security rules, Storage, Functions and Hosting, and finally offline support, emulators and protecting the project. The same collections, field names and user ids appear in every lesson, so you are never learning a new example and a new idea at the same time. Simple figures, such as the StudyCircle data tree or the flow of a rules check, appear where they help. You go at your own pace.
MCQ practice vs video courses for Firebase
Video courses on Udemy, Coursera or YouTube are a good choice if you want to watch someone click through the Firebase console and build an app from scratch. This course does a different job: a short explanation followed straight away by questions about real code and rules, such as whether updateDoc fails on a missing document, whether a query will be rejected by the rules, which query needs a composite index, or what happens to subcollections when a document is deleted. Firebase behaviour is precise, and predicting it question by question builds the understanding that keeps production data safe. Many developers use both: a video for a first walkthrough, and questions to make the details stick.
Common Firebase mistakes
The course corrects these directly:
- Hiding the web API key. The key in the config object identifies the project and is not a secret; security rules and App Check control access.
- Treating security rules as filters. A query is rejected if it could return any document the user may not read, so the query must match the rules.
- Shipping test mode rules. They allow all reads and writes and expire, leaving an app either open or broken.
- Expecting deleteDoc to remove everything. Deleting a document leaves its subcollections behind.
- Forgetting to unsubscribe from onSnapshot listeners, which keeps reads running after a page is gone.
- Using sequential document ids, which creates write hotspots.
- Writing Cloud Functions that are not idempotent. Firestore triggers are delivered at least once, so a handler may run twice.
Best way to learn Firebase
Model the data around the queries your screens need, then write the rules for those queries before you write the screens. Run everything against the Local Emulator Suite so you can test rules and functions without touching a live project. The last part of the course works this way: it compares Firestore with Realtime Database and with a SQL database, asks when Firestore fits an app, when to add Cloud Functions and when to block client-side writes, and applies these choices to a chat app with real-time listeners, an admin panel with custom claims, image uploads with storage triggers and a multi-tenant SaaS data model.