Overview
We are seeking a contract Machine Learning Engineer specializing in recommender systems, who will play a crucial role in building and deploying advanced recommendation, ranking, and retrieval models in production environments. The successful candidate will collaborate with cross-functional teams to implement real-time and batch personalization features, including personalized search and next-best-action recommendations. This role can be performed remotely or in a hybrid setup across the UK, with the initial contract lasting 3-6 months and potential extensions.
Responsibilities
- Build and deploy recommendation and ranking models in production.
- Develop candidate generation and re-ranking pipelines at scale.
- Implement real-time and batch personalization features.
- Conduct offline evaluations and online A/B testing.
- Collaborate with cross-functional teams to gather requirements.
- Optimize models for performance and accuracy.
Requirements
- Proven experience in building production recommender systems.
- Strong proficiency in Python and frameworks like PyTorch or TensorFlow.
- Familiarity with recommender techniques such as collaborative filtering and matrix factorisation.
- Experience with embeddings and approximate nearest-neighbour search technologies.
- Proficient in Spark or PySpark for feature engineering at scale.
- Experience with real-time data processing technologies like Kafka or Kinesis.
- Knowledge of model serving methodologies and frameworks.
- Familiarity with cloud services such as AWS, GCP, or Azure.