Join our ML team and work on the algorithms and systems that sit at the heart of our product. As a Machine Learning Engineer, you'll take ML models from research prototype to robust production deployment, building the infrastructure and tooling that makes our models reliable, observable, and continuously improving. This is a role for someone who bridges the worlds of research and software engineering.
What you'll do
- Train, evaluate, and deploy machine learning models using PyTorch, scikit-learn, or similar frameworks into production serving infrastructure.
- Build MLOps pipelines covering data preprocessing, model training, evaluation, versioning, and A/B testing of new model versions.
- Design low-latency inference services that meet p99 SLA requirements under production traffic loads.
- Collaborate with data scientists to productionize research models, bridging the gap between experimentation and reliable deployment.
- Monitor model health in production, detecting drift, degradation, and data quality issues before they impact user experience.
What we're looking for
- Three or more years of experience deploying machine learning models to production environments serving real traffic.
- Proficiency in Python and common ML libraries including PyTorch, TensorFlow, or JAX.
- Experience with feature stores, model registries, and ML experiment tracking tools like MLflow or Weights & Biases.
- Strong software engineering skills — you write clean, testable, documented code and think carefully about system design.
- Familiarity with vector databases, embedding models, and retrieval-augmented generation (RAG) architectures is a plus.
We offer highly competitive total compensation with strong equity, comprehensive benefits, access to significant compute resources for experimentation, conference and publication support, and a team culture that respects deep technical work. You'll collaborate with world-class researchers and engineers on problems that genuinely push the field forward.