Shuling technology

Design and implement an event-level media recommendation engine.

Build candidate recall based on event, venue, time phase, GPS range, media status, and audience type.

Implement visual similarity recall using image embeddings such as SigLIP.

Work with vector search engines such as Qdrant.

Design ranking scores based on freshness, geo distance, event phase, image quality, popularity, and user behavior.

Build recommendation logs for impression, click, dwell time, like, save, share, download, and group join.

Help design attribution analysis and feedback loops for continuous recommendation improvement.

Integrate the recommendation service with our Node.js backend and Flutter app.

Build APIs for feed ranking, similar media search, media upsert, and interaction tracking.

Help us move from rule-based ranking to data-driven ranking as usage grows.

We are building HypeMuse, an event-based photo and GIF discovery platform for sports games, concerts, and live events.

Our current backend is built with Laravel/PHP, and we are planning to rebuild the backend using Node.js. We are looking for a strong backend engineer who can help migrate the existing Laravel system into a scalable Node.js architecture while keeping the current product logic stable.

Responsibilities:

- Analyze the existing Laravel backend structure, database schema, APIs, and admin workflows.

- Rebuild core backend services using Node.js, preferably TypeScript with NestJS or Express.

- Design clean REST APIs for the mobile app, admin panel, media upload, user accounts, events, venues, and recommendation feed.

- Integrate with existing services including MySQL/PostgreSQL, Redis, object storage, and the recommendation service.

- Support event-based recommendation logic for photos and GIFs, including event ID, venue location, match phase, audience type, media quality, freshness, and popularity.

- Build background jobs for media processing, metadata sync, recommendation data sync, and interaction tracking.

- Set up Docker-based deployment and production-ready environment configuration.

- Improve backend performance, security, logging, monitoring, and maintainability.

- Work closely with the founder/product owner to turn the current Laravel backend into a scalable Node.js backend.

Requirements:

- Strong experience with Node.js backend development.

- Good knowledge of TypeScript, NestJS or Express.js.

- Experience working with Laravel/PHP projects and backend migration is highly preferred.

- Strong understanding of REST API design, authentication, database modeling, and queue-based systems.

- Experience with MySQL or PostgreSQL, Redis, Docker, and Linux server deployment.

- Ability to read an existing codebase and migrate features without breaking current product behavior.

- Experience with media platforms, recommendation systems, feeds, image/video processing, or event-based products is a plus.

Nice to have:

- Experience with Qdrant, Faiss, vector search, recommendation systems, or AI-powered media tagging.

- Experience with cloud storage, CDN, image optimization, and large-scale media upload.

- Experience with sports, concert, creator, or community platforms.

This role is suitable for someone who can work independently, communicate clearly, and help us move from a Laravel MVP backend to a more scalable Node.js backend architecture.

Basic event feed backend structure

2. Media item status model

3. Redis cache workflow for event feed

4. Basic Fast Lane upload processing flow

5. trigger_type and highlight_triggers backend structure

6. One sample admin-triggered workflow, for example goal or full-time

7. Status transition from provisional_ready to ready / pending_review / blocked