Building for Your People, Not for Everyone: Why I'm Rethinking Community Platforms

I spent the better part of last year watching my team's internal Slack channel turn into a graveyard of half-finished conversations. Someone would post a legitimate technical debate, we'd exchange three messages, and then—dead. The context would get buried under notifications, someone would ask the same question two weeks later, and we'd repeat the cycle. I kept thinking: why doesn't there exist a space where actual discussion lives, where disagreement isn't performance art for an algorithm?

Then I read about Sajha, and it clicked. This isn't just another forum. This is someone saying "the tools we inherited don't fit how we actually talk," and building specifically for that gap.

When Off-the-Shelf Doesn't Fit

The original problem Yukesh encountered is one I recognize from my own work. You've got Twitter/X where everything's a hot take performance. You've got Reddit, which works, but it's designed for English-speaking global audiences with specific cultural references. You've got Facebook which is basically a family reunion you can't escape. None of these feel like home if you're building community with specific people in a specific place.

What impressed me about the Sajha approach wasn't the technical complexity—it was the specificity of the problem-solving. Instead of building another generic upvote/downvote system, he renamed it. Flame it. Cool it. That's not semantic game-playing; that's acknowledging how conversations actually work in certain communities.

The Architecture Choice That Actually Matters

Here's where I had to sit with my own biases. WordPress as the backend for a modern community platform? My first instinct was the same one I always have: that's not a "serious" technical choice. But then I realized I was falling into the exact trap the article warns against.

Yukesh made a deliberate call: iterate fast on the actual product problem—will people use this?—rather than optimize for resume-building technology choices. WordPress meant he could focus on the custom frontend layer, the voting mechanic, the community structure. The unsexy foundation handled authentication, moderation tooling, and persistence so he didn't have to reinvent them from scratch.

I've built systems the other way. From-scratch React frontend, Node backend, PostgreSQL database, Redis cache. It was technically beautiful and horizontally scalable and absolutely overkill for testing whether the core idea had merit. What Sajha teaches is: know what problem needs your custom engineering and what problem needs someone else's solved problem.

The Feature Ruthlessness I Actually Respect

The part of this that genuinely stuck with me is what he didn't build. No karma system. No elaborate notification engine. No messaging feature. He cut these not because they're hard, but because they weren't questions he needed answered yet.

I'm guilty of the opposite. I'll build features because they sound necessary, because other platforms have them, because future users might need them. I've shipping notification systems that nobody uses. I've built admin dashboards that were never opened. The discipline to ask "does this answer my core question?" and cut ruthlessly—that's expertise.

// The voting mechanic—simple but intentional
const handleVote = (postId, voteType) => {
  if (voteType === 'flame') {
    incrementFlameCount(postId);
    logUserVote(userId, postId, 'fire');
  } else if (voteType === 'cool') {
    incrementCoolCount(postId);
    logUserVote(userId, postId, 'sits_down');
  }
  // No karma calculations. No notification queue.
  // Just: did this resonate with you?
};

Notice what's not in that function: no reputation score updates, no algorithmic weighting, no dark-pattern engagement tracking. It's serving the interaction, not the metrics.

My Real Questions

What I'm still sitting with: How does Sajha maintain signal without algorithmic amplification? What prevents it from becoming either a ghost town or a cesspool? Community moderation is where most platforms actually fail, and I'd genuinely want to know how they're approaching governance at scale.

Also—and this might be my Islamabad perspective showing—regional platforms feel important right now. We have different conversations than Western tech Twitter. We have different concerns. But building something sustainable that isn't venture-backed, that doesn't need to exit to a billion-dollar acquirer, is its own kind of hard that I don't think gets discussed enough.

If you're annoyed enough at the existing options to build your own alternative, that's usually the signal that you're solving a real problem. I'm curious to see if Sajha actually sustains that annoyed energy into something durable.

Source: This post was inspired by "How I Made Sajha (and why I think Nepal needed it)" by Dev.to. Read the original article