What is Firebase?

Firebase is Google's development platform that bundles the backend services most applications need (database, authentication, file storage, hosting, and analytics) into a single, integrated console. Teams spend less time wiring together separate providers and more time building product.

What Firebase provides

Building an application from scratch normally means choosing a database, configuring an authentication system, arranging file storage, setting up hosting, and adding monitoring, each from a different vendor. Firebase replaces that patchwork with a single ecosystem where every service shares the same configuration, security rules, and management console. It is a backend as a service platform, meaning Google operates and scales the underlying infrastructure for you.

All Firebase services connect to Google Cloud, so teams that outgrow the platform's built-in limits can reach BigQuery for advanced analytics, custom machine learning infrastructure, or broader Google Cloud tooling without migrating away entirely.

Database options

Firebase offers two distinct approaches to data storage.

Realtime Database is Firebase's original offering. It stores data as JSON and pushes updates to every connected client the moment data changes, making it well suited to chat features, collaborative editing, and live dashboards.

Cloud Firestore is the more recent, document-based option. It supports structured collections, complex queries, offline access, and smoother horizontal scaling. Most new projects start here unless instant synchronisation at minimal cost is the primary requirement.

For a detailed comparison with another popular platform that offers similar choices, see the difference between Firebase and Supabase.

Core services

Authentication

Firebase Authentication handles the full user identity flow: email and password, phone verification, and social logins from Google, Facebook, Twitter, and GitHub. It manages account creation, password resets, and token lifecycle, and integrates directly with Firestore and Storage security rules so access control stays consistent across the stack.

Storage

Firebase Storage is built on Google Cloud Storage and provides scalable hosting for user-generated content such as images, videos, and documents. Access control ties directly into Authentication, so rules like "only the uploading user can delete their file" require a few lines of configuration rather than custom server code.

Hosting

Firebase Hosting delivers web applications through a global CDN with free SSL certificates and straightforward custom domain support. It integrates with modern frameworks and supports server-side rendering via Cloud Functions.

Cloud Functions

Cloud Functions run server-side code in response to Firebase events or HTTP requests without requiring a managed server. Common uses include sending notifications, processing payments, and automating data transformations. Firebase handles scaling automatically.

Cloud Messaging

Firebase Cloud Messaging (FCM) enables push notifications across iOS, Android, and web. Teams can target specific audience segments, schedule deliveries, and run A/B tests on message content.

Analytics

Firebase Analytics tracks user behaviour, conversion events, and retention. Because it shares the same project configuration as the rest of Firebase, it can be linked directly to Authentication segments, remote config experiments, and A/B tests within the console.

How Firebase compares to similar tools

Firebase is often evaluated alongside Supabase, which takes a Postgres-first, open-source approach. The core trade-offs are Firestore's document model and real-time sync versus Supabase's relational queries and SQL familiarity. Both are backend-as-a-service platforms; the right choice depends on data shape, query complexity, and whether vendor lock-in is a concern.

For a broader view of where Firebase sits in the backend-as-a-service landscape, see what is backend as a service.

In an AI-native team

Coding agents can scaffold a Firebase project (initialising collections, writing security rules, and generating Cloud Functions) in minutes from a short prompt. The practical risk is that generated security rules can look plausible while containing privilege escalation paths, so engineers still need to read and verify every rule before deployment. Teams using agentic coding with Firebase benefit most when someone on the team understands Firestore's data model and security rule evaluation deeply enough to review output that they did not write themselves, following the discipline described in how to review AI-generated code.

What we test for

When placing engineers who will work with Firebase, we assess both database design judgment and security rule literacy, since mistakes in either compound quickly at scale. Our how we vet process runs two sessions: a fundamentals assessment without AI tools, where candidates design data models and write security rules by hand, and an AI-native assessment where they use coding agents and must verify the generated configuration, catching errors in code they did not author (the skill set described in testing software you didn't design).

Short answers