Welcome to one of the most exciting lessons in this course, Reputation and Recommendation.

You might be wondering — aren’t these two totally different things? Why are we talking about them together? Well, not really. Let me explain.

Both reputation and recommendation systems are built on one core thing: user interaction.

Whenever a user does something on your platform—whether it’s viewing a question, posting an answer, or upvoting—that action tells you something. Either you reward it with points (reputation), or you log it to learn what the user likes (recommendations).

For example, If a user checks out a question detail page, that probably means they’re interested in that topic.

• If they upvote something — that’s a signal.
• If they downvote — that’s a signal, too.
• They post a question? Another signal.
• They answer one? Yup, you guessed it — another one..

No, no, this isn’t a Mark Zuckerberg "we track everything" kind of situation — just kidding 😄

The point is that you keep track of useful user actions and reward them with points—that’s how you build their reputation.

• Post a question → +5 points
• Answer a question → +10 points
• Upvote → +2 points
• Downvote → -2 points

And for everything else users do, you gather that data, analyze it, and start spotting patterns:

• Are they always answering JavaScript questions?
• Do they love anything with “Next.js”?
• Are they big on PHP?

You study these patterns — and that’s how you build a recommendation system.

Not magic. Not rocket science. Just smart use of data.

Of course, as your platform grows and your data gets bigger and messier, you’ll need a more advanced system and logic, but the foundation stays the same: Track user interactions, analyze them well, and use them to reward and recommend.

Alright, let’s dive in 🚀