Machine Learning Engineer, Retrieval

About the role

Teams often tune a model for weeks and leave the index alone, when the index was the constraint the whole time. This role exists because retrieval quality is the work with the biggest payoff in applied AI, and few teams staff it on purpose. You will own chunking, embedding, ranking, hybrid search and the measurement that proves any of it helped.

What you will do

• Build and tune retrieval over client documents and data that were never designed to be searched.
• Run offline evaluations that predict production behaviour, and say so when they do not.
• Work on ranking: hybrid keyword and vector search, reranking, and knowing when a simpler approach wins.
• Instrument retrieval in production so a regression is visible before a client reports it.
• Share what you learn with the team, and now and then on our blog.

What we need from you

• Practical experience with embeddings, vector search and the failure modes of both.
• You have measured a retrieval system rather than assuming it worked.
• Solid Python and SQL. You are comfortable inside a database, not only above it.
• An honest relationship with metrics: you know what recall at k does and does not tell you.

Useful but not required

• An information retrieval background, formal or self-taught.
• Experience with multilingual or heavily domain-specific content.
• You have built a reranker and can say whether it was worth it.