Trendy

Turn trending topics into publish-ready content in hours, not days, using AI-powered research, planning, review, and workflow automation.

Comment

Trendy

Inspiration

Creating high-quality content is slow. By the time creators finish researching trends, writing articles, editing, and publishing, the conversation has often already moved on.

We built Trendy to solve this problem. Instead of humans chasing trends, autonomous AI agents continuously monitor the internet, identify high-value topics, generate SEO-optimized content, and publish it automatically. The goal is to transform content creation from a manual workflow into an autonomous pipeline.

What it does

Trendy is an AI-powered autonomous content engine that discovers trending topics, generates publish-ready articles, and publishes them with minimal human intervention.

The platform coordinates five specialized AI agents:

  1. Trend Scorer

    • Monitors Reddit, Google Trends, and Twitter/X.
    • Scores trending topics based on popularity and publishing opportunity.

  2. Brief Generator

    • Converts selected topics into detailed SEO-focused content briefs.

  3. Workload Matcher

    • Assigns briefs to virtual writer profiles based on expertise and availability.

  4. Copywriter & Reviewer

    • Researches the topic.
    • Produces a complete Markdown article.
    • Reviews readability, SEO quality, and formatting.

  5. Publisher

    • Validates metadata and Markdown.
    • Publishes directly to platforms such as Dev.to.

A real-time dashboard lets operators monitor every agent through a live terminal-style log stream.

How we built it

Frontend

  • React 19
  • TypeScript
  • Vite
  • React Router DOM 7
  • Custom Glassmorphism UI

Backend

  • Node.js
  • Express.js
  • Prisma ORM

Database

  • PostgreSQL
  • Supabase

AI

  • Qwen LLMs powering:
    • Trend analysis
    • Brief generation
    • Content writing
    • Quality evaluation

Integrations

  • Google Trends
  • Reddit API
  • Twitter/X API
  • Dev.to Publishing API

Challenges we ran into

Real-time streaming

Building a responsive terminal that streams backend agent logs in real time without blocking the UI.

Concurrent database workloads

Multiple AI agents executing simultaneously created connection bottlenecks. We optimized Prisma connection pooling and tuned our PostgreSQL configuration on Supabase.

Publishing compatibility

Ensuring AI-generated Markdown consistently met Dev.to's formatting and metadata requirements without malformed output.

Accomplishments that we're proud of

  • Built a complete autonomous five-agent workflow from trend discovery to publishing.
  • Designed a modern glassmorphic dashboard with live agent monitoring.
  • Created reliable coordination between independent AI agents through validation and handoff stages.
  • Reduced manual content production into a single automated pipeline.

What we learned

  • Designing collaborative AI agent systems instead of relying on one large prompt.
  • Streaming backend events efficiently to a modern React frontend.
  • Managing concurrent Prisma database operations under heavy AI workloads.
  • Structuring AI-generated content for production publishing platforms.

What's next for Trendy

  • Support publishing to WordPress, Ghost, Medium, Hashnode, and custom webhooks.
  • Add post-publication analytics that feed engagement metrics back into the Trend Scorer.
  • Automatically generate cover images and article illustrations using AI.
  • Introduce scheduling, approval workflows, and collaborative editing for content teams.
  • Expand trend discovery to YouTube, Hacker News, Product Hunt, and LinkedIn.

Built With

Share this project:

Updates