Machine Learning Engineer, Ranking & Retrieval

About

There are over 5 billion users using basic applications today such email, notes, tasks, calendar and they're not AI-native. Our mission is to build proactive applications for anyone in the world, who are not used to complex prompting. We aim to bring intelligence to conversations, errands, organising and workflows, with minimal to no prompting.

Our first application is an agent-first mobile communication app. We believe building applications that users use daily for communication, coordination and collaboration needs will help integrate AI into their lives. The net effect is to eliminate misinformation, achieve low coordination costs, enable high quality tasks and collaboration at scale with minimal or little prompting.

Role

- Develop and improve ML models for classification, retrieval, ranking and personalisation
- Build classifiers and ranking models, optimising accuracy, precision, latency and inference cost
- Improve search relevance and context retrieval using embeddings, hybrid search and reranking
- Develop personalisation models using user behaviour and product data
- Build evaluation pipelines, datasets and metrics to measure model quality and detect regressions
- Train, fine-tune and deploy models using production ML pipelines
- Monitor model performance, debug production issues and improve models based on evaluation results and user feedback

Requirements

3+ years of machine learning engineering experience, including deploying models to production; Strong Python programming skills and experience with PyTorch or JAX; Solid understanding of machine learning algorithms, model training and evaluation; Experience with classification, ranking, retrieval, recommendation or personalisation models; Experience working with embeddings, vector search or information retrieval systems; Experience building data pipelines, training workflows and model inference services; Ability to evaluate model performance using relevant metrics and experimentation; Understanding of trade-offs between model accuracy, latency, computational cost and reliability

Nice to have

Experience with transformer models, LLM fine-tuning or embedding models; Experience with hybrid search, reranking and retrieval evaluation; Experience with recommendation systems or personalisation in consumer applications; Experience with large-scale ML training or inference infrastructure