Skip to content

Repository files navigation

Align Network

Aligning minds, just for you. A social network where creators find their people based on how they think, not follower counts.


What is Align?

Align is a creator platform built on a simple idea: social networks shouldn't reward virality—they should reward thinking.

On Twitter, you get algorithmic amplification for rage. On Instagram, it's aesthetics. On LinkedIn, it's job titles. None of them help you actually find your people.

Align does. We connect creators with audiences that genuinely care about their work—through shared interests, verification, and transparent algorithms.

The Problem We're Solving

Every major social platform is fundamentally broken for creators who care about depth:

  • Engagement gaming: Posts that trigger emotion outrank posts that create value
  • Follower-based visibility: Being new means invisibility, regardless of quality
  • Algorithm opacity: You post, something happens, you don't know why
  • Network lock-in: You're trapped following people you originally followed for one reason

Result: Creators either compromise their work or quit.

Our Solution

An algorithm designed from the ground up for thinking alignment:

  1. Match by interests, not virality — Posts surface based on how well they align with what you actually care about
  2. Verify quality creators — Verified creators float to the top, but the algorithm still discovers new talent
  3. Weight engagement by type — Comments matter more than likes (they indicate real engagement)
  4. Show your reasoning — Every post explains why it appeared in your feed
  5. Kill the cold start problem — New creators with good ideas still get seen

How It Works

The Algorithm (Simplified)

Every post gets scored on:

Score =
  0.40 × (You follow this creator) +
  0.25 × (Post matches your interests) +
  0.20 × (Quality engagement: comments/bookmarks vs. likes) +
  0.15 × (Creator's audience size—mild boost) +
  0.10 × (Post is recent) +
  0.10 × (Creator thinks like you)

Why this works:

  • Followers matter (0.40) because you consciously chose to follow for a reason
  • Interest matching (0.25) enables discovery beyond your network
  • Engagement quality (0.20) means thoughtful posts beat popular posts
  • Social proof (0.15) surfaces established creators without letting them takeover
  • Recency (0.10) keeps feeds fresh without burying old quality content
  • Similar thinking (0.10) adds serendipity—discover people exploring adjacent ideas

What You See

Every post in your feed shows:

  • Why it appeared: "Matches: TypeScript, System Design" or "From @designer.jane"
  • How much weight each signal had: Transparency builds trust
  • Comments prioritized over likes: The algorithm rewards depth

Key Ideas

1. Interest-Based Matching (Not Following-Only)

Your feed isn't limited to who you follow. Instead:

  • You tell us your interests: "TypeScript", "Design Systems", "Cloud Architecture"
  • Posts get automatically tagged with topics
  • We use Jaccard similarity to match you with relevant posts, even from creators you don't follow yet

This solves the discovery problem—new creators with good ideas don't get buried.

2. Verification Prioritizes Without Gatekeeping

Verified creators (curated by us) float to the top, but:

  • Non-verified creators still rank by the full algorithm
  • New creators can still break through with good content
  • Verification is earned, not purchased

This prevents the "Twitter effect" where only celebs get visibility.

3. Engagement Quality Over Volume

We differentiate between engagement types:

  • Likes: Passive, weighted lowest
  • Comments: Active engagement, weighted higher
  • Bookmarks: "I want to reference this later", weighted highest

A post with 10 comments beats a post with 100 likes because comments indicate real thought.

4. Transparent Reasoning

When you see a post, you see:

  • Why the algorithm showed it to you
  • How much each factor influenced the decision
  • The actual scoring breakdown

This builds trust. You're not confused by algorithmic magic—you understand why you're seeing what you're seeing.

5. Creator-Friendly Discovery

For creators:

  • Your posts surface based on how well they match audience interests
  • No need to game engagement metrics
  • Quality thinking gets rewarded over sensationalism
  • New creators can break through without viral mechanics

Platform Features

Feed

  • For You: Interest-based discovery + your follows
  • Following: Only people you explicitly follow
  • Personalized: Changes based on your interests and engagement

Who to Follow

  • Recommendations based on:
    • Mutual connections
    • Shared interests
    • Network overlap
  • Shows why we're suggesting each person

Creator Profiles

  • Bio, interests, verification badge
  • Post history with engagement metrics
  • Follower/Following lists
  • Badges: Verified, Early Adopter, Active Creator

Communities

  • Interest-based discussion spaces
  • Community creators set the culture
  • Separate feed within each community
  • Moderation by community moderators

Bookmarks

  • Save posts privately
  • Organize into collections
  • No algorithm interference—just what you saved

Search

  • Find creators by name, interest, or bio
  • Find posts by keywords or hashtags
  • Ranked by relevance (interest match + engagement)

Why Align Will Win

For Creators

✅ Build audience around thinking, not gaming
✅ New creators can be discovered without virality
✅ Transparent why posts succeed or fail
✅ Community is genuinely interested in your work

For Users

✅ See content that actually matters to you
✅ Discover people thinking about what you care about
✅ No rage-bait or engagement manipulation
✅ Understand why you're seeing what you're seeing

For the Internet

✅ Prove social networks can exist without engagement gaming
✅ Show verification + algorithm can coexist
✅ Build incentive structures that reward depth


The Thinking Behind Decisions

Why 0.40 weight for followers?

Followers are intentional signals. Users consciously chose to follow this creator. But 0.40 (not 0.6+) ensures discovery beyond your current network—you want both the people you follow AND people exploring your interests.

Why Jaccard similarity for interests?

Set-based similarity handles categorical data well. When we say "matches 3 of 5 interests," users immediately understand why. No black-box embeddings—just honest math.

Why verified creators get priority but not absolute dominance?

Verification should prioritize quality without gatekeeping. Verified posts float to top, but we still rank them by recency within that tier. Non-verified creators still rank by full algorithm. This balance encourages quality without crushing new talent.

Why limit to 3 posts per author?

Feed diversity is proven to reduce filter bubbles and improve engagement. Three posts lets prolific creators maintain visibility without feed takeover.

Why weight comments 3x likes?

Comments require thought. Someone who comments engaged with your idea, disagreed, added insight, or discussed. Likes are passive thumb reactions. Comments are effort.


What Makes This Different

Platform Optimizes For Consequence
Twitter Engagement & time-on-site Rage dominates
Instagram Aesthetic appeal Substance loses
LinkedIn Perceived status Fakeness spreads
TikTok Watch time Content quality suffers
Align Thinking alignment Quality creators thrive

The Vision

We're building a social network that treats creators and users like intelligent people. No engagement tricks. No viral loops designed to addict. No algorithms you can't understand.

Just: creators making their best work, finding people who actually care, building real communities.

That's Align.


Get Involved

Join the Community

Built By

Shrvan — Founder & Developer
Building Align because existing platforms are broken.


License

MIT — Build with it, improve it, learn from it.


The creator economy doesn't need another engagement machine. It needs alignment.

About

a chill corner for creative kids to build & ship

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages