Skip to main content

ClickUp

Machine Learning Engineer, Ranking & Retrieval

United States (Remote)$200k–$250kfulltimemidAdded 3 days ago

About this role

ClickUp seeks an experienced Machine Learning Engineer to own the full lifecycle of ranking and retrieval systems powering search across their AI-native workspace platform. You'll build and deploy models that surface relevant, permissions-aware results at massive scale across billions of documents.

What you'll do

  • Train, deploy, and serve ranking models in production with end-to-end ownership
  • Build ranker features, training pipelines, and offline evaluation frameworks
  • Design hybrid retrieval systems combining lexical and vector search with HNSW indexing
  • Execute embedding inference at billion-scale document volumes
  • Improve query understanding through intent modeling and query expansion
  • Create measurement frameworks to evaluate and continuously improve search quality

What they're looking for

  • Ranking and retrieval systems (5+ years)
  • Full ML lifecycle: training, deployment, production serving
  • Hybrid retrieval (lexical + vector search)
  • Large-scale embedding inference
  • Query understanding and NLP fundamentals
  • OpenSearch or Elasticsearch
  • Feature engineering and offline evaluation
  • Python or similar ML languages
Apply with Autofill

Opens the application — the Jobs AI extension fills it for you. Set up autofill

Opens the official application on the employer’s site. No login required.

ClickUp

ClickUp builds a work management and productivity platform while investing in AI-powered automation and integration capabilities across its go-to-market and business systems infrastructure. The company is hiring for DevOps engineers, solutions engineers, and business systems engineers to support cloud infrastructure, customer implementations, and intelligent workflow automation.

View all jobs at ClickUp

Likely interview questions

  • Walk us through a ranking model you've trained from feature engineering through production deployment—what metrics did you optimize for?
  • How have you balanced lexical and vector search in a hybrid retrieval system, and what trade-offs did you encounter at scale?