SOC-0033: Develop a Custom Food Feed for Bluesky

Project size: 3 months

Estimated difficulty: Intermediate

Recommended Skills

• Proficiency in TypeScript and Deno
• Familiarity with Bluesky’s AT Protocol and Jetstream
• Experience publishing libraries to the JavaScript Registry (JSR)
• Knowledge of content filtering and recommendation algorithms
• Experience deploying and scaling Deno apps on Fly.io

Description

Bluesky’s decentralized social network offers open, federated content streams but lacks tailored feeds for niche interests like culinary content. This project will build a full‐stack pipeline to aggregate, filter, rank, and serve only food-related posts:

1. Publish a Deno Jetstream client library to JSR, enabling easy consumption of Bluesky feeds in Deno projects.
2. Implement feed filters that isolate recipes, restaurant reviews, cooking tips, and food photography.
3. Design and apply sorting algorithms to surface top posts by relevance, engagement metrics, and quality signals.
4. Create a local test suite to validate end-to-end functionality under varying load and data scenarios.
5. Deploy the pipeline on Fly.io with CI/CD, monitoring, and auto-scaling for reliable, real-time delivery.

By project end, food enthusiasts will enjoy a live, curated Bluesky feed showcasing the best culinary content from around the world.

Expected Results

• Published Deno Jetstream library on JSR with comprehensive README and usage examples
• Robust filter modules that accurately detect and extract food-related posts
• Ranking engine with configurable scoring based on engagement and content quality
• Local test harness including sample scripts and automated tests
• Fly.io deployment config and documentation for continuous integration and scaling
• Live custom feed endpoint serving real-time, curated food content on Bluesky