Machine Learning Engineer - Search Systems

Required for a 12-month initial onsite in Lund, Sweden, an ML Engineer to join a team responsible for a large-scale, production search system that combines lexical retrieval (Solr/Lucene) with semantic and ML-based ranking.

Role:

To understand why search models perform the way they do, and drive measurable improvement.

You'll work across the full stack of a mature search system: analysing ranking behaviour and query/document signals, designing and running experiments, and shipping changes into a high-throughput, low-latency production environment built primarily in Java.

Responsibilities:
- Diagnose root causes of under-performance in existing ranking/relevance models.
- Design, validate, and ship new signals and features.
- Improve the integration between lexical (Solr/Lucene) and semantic/ML retrieval and ranking.
- Deliver changes into a primarily Java-based production search stack.

Must-have Qualifications:
- Strong programming and software engineering fundamentals (language-agnostic; JVM familiarity).
- Track record improving/debugging existing, already-tuned ML models.
- Solid grounding in classical ML and ranking techniques.
- Hands-on experience with lexical search (Solr/Lucene/Elasticsearch) combined with semantic/vector search.
- Practical experience with ML frameworks (PyTorch, TensorFlow, Scikit-learn, NumPy, Pandas).
- Strong experimental rigor: designing tests, reading ranking metrics, separating signal from noise.

To arrange a Teams-based interview, please email in the first instance, your CV to (see below)

WA Consultants is an Employment Business and an Employment Agency as described within The Conduct of Employment Agencies and Employment Businesses Regulations 2003.