Data Scientist / Machine Learning Engineer (Recommendation Systems)

Job Description

Job Description

About Beatdapp

Beatdapp is a venture-backed startup delivering the most advanced streaming integrity and recommendation technology in the world1. While our roots are in fighting the multi-billion dollar problem of streaming fraud, we have leveraged our "Trust & Safety Operating System" to power a new generation of discovery.

We believe that true personalization starts with verified behavior. By filtering out noise and manipulated signals before they impact the model, we build recommendation engines on a foundation of clean, authentic data. We are looking for builders who want to work with the world's best streaming services and music labels to reshape how content is discovered.

The Role

We are seeking a Data Scientist / Machine Learning Engineer who specializes in Recommendation Systems. You will move beyond simple analysis to research, build, and deploy production-grade models that power discovery for millions of users.

In this role, you will bridge the gap between research and engineering. You will design advanced architectures and ensure they run efficiently at scale. You will work closely with our leadership and data engineering teams to turn trillions of data points into experiences that anticipate user intent in real-time.

Responsibilities

• Build End-to-End Recommendation Pipelines : Design and implement scalable recommendation architectures to surface relevant content from large catalogs.
• Full-Cycle Project Ownership : Take ownership of projects across the complete Machine Learning lifecycle, driving initiatives from initial problem formulation and exploratory analysis to model training, validation, and post-deployment monitoring.
• Advanced Behavioral Modeling : Develop and train deep learning models (e.g., GNNs, Transformers, Wide-to-Narrow networks) to create rich user and item embeddings based on authentic interactions.
• Scalable Multi-GPU Training : Design and execute distributed training workflows for large-scale deep learning models, utilizing multi-GPU strategies and parallel computing techniques to maximize training throughput and handle massive datasets efficiently.
• Strategic Signal Extraction & Feature Modeling : Systematically mine and sift through high-dimensional user, track, and streaming event data to distinguish between subtle implicit signals and explicit feedback, mathematically modeling these behaviors to engineer dense, predictive features that enhance model performance.
• Production Engineering : Write clean, production-ready code (Python) and oversee the deployment of models into high-availability environments. You will optimize models for low latency to ensure instant load times.
• Cross-Functional Collaboration : Partner closely with Product and Engineering teams to translate business requirements into technical specifications, ensuring seamless development, integration, and deployment of models into the core product ecosystem.

Successful Candidates will have

Bonus Points