Machine Learning Engineer – Matching & Recommendations

About the role

At Bumble, the Machine Learning Engineer for Matching & Recommendations will develop and improve the ML systems that power matching, recommendation and personalisation experiences for our members. You will own well‑defined ML problems end‑to‑end, from data exploration through deployment and monitoring, while collaborating with product, engineering and data teams.

Key responsibilities

• Explore, develop and deliver modern ML solutions that improve recommendations, matching and personalisation across Bumble.
• Own defined ML problems end‑to‑end, from data exploration and feature engineering through model training, evaluation, production deployment and iteration.
• Use modern ML frameworks such as PyTorch or TensorFlow to design, train and optimise models for production environments.
• Contribute to experimentation, including A/B testing and offline evaluation, using results to continuously improve model and product performance.
• Build, maintain and monitor production models, diagnosing issues and improving reliability and performance at scale.

Required profile

• ~3+ years of hands‑on experience building and shipping ML models in production.
• Strong programming skills in Python and proficiency with an ML framework such as PyTorch or TensorFlow.
• Experience with recommendation systems, ranking, retrieval or personalisation.
• Good understanding of the ML development lifecycle, from data and feature development to monitoring and iteration.
• Knowledge of MLOps concepts such as CI/CD for ML, feature stores, model serving, observability and versioning.
• Familiarity with containerisation and cloud‑native environments like Docker, Kubernetes and GCP.
• Experience with experimentation methodologies, including A/B testing and offline evaluation.

Required skills

• Python
• PyTorch
• TensorFlow
• Docker
• Kubernetes
• Google Cloud Platform (GCP)
• CI/CD for ML
• Feature stores
• Model serving
• Observability and versioning
• A/B testing
• Offline evaluation