Machine Learning Engineer — Boston, MA
Зарплата не указана
ОткликнутьсяAbout The Role
The role is a hands-on ML engineering position focused on taking models from notebook to production: training pipelines, deployment infrastructure, latency-sensitive serving, and continuous evaluation at scale.
The team ships ML systems that customers depend on daily — this role sits at the intersection of applied science and backend engineering, and its work directly shapes product quality and revenue impact.
Key Responsibilities
• Design, train, and evaluate production ML models (classification, ranking, NLP, forecasting) that directly drive core product decisions
• Build and maintain end-to-end training and feature pipelines in Python and PySpark, ensuring data quality and reproducibility across environments
• Deploy and operate models on cloud infrastructure (AWS SageMaker, GCP Vertex Fr, or Kubernetes-based serving) with versioning, canary releases, and rollback procedures
• Instrument monitoring for data drift, prediction skew, and performance regression; own alerting and incident response for production models
• Run structured experiments and A/B tests; translate offline metrics into validated online business impact
• Collaborate with data scientists and backend engineers to define model APIs, latency budgets, and serving architectures
• Contribute to ML platform tooling, code reviews, and internal engineering standards
What We Are Looking For
• 3–6 years of experience in machine learning engineering, applied ML, or equivalent production-focused data science work
• Track record of shipping at least 2–3 models to production, including ownership of serving and monitoring
• Strong Python engineering skills plus deep proficiency with PyTorch, TensorFlow, or scikit-learn
• Hands-on experience with cloud ML platforms (SageMaker, Vertex Fr, Azure ML) or containerized model serving on Kubernetes
• Solid ML fundamentals: model selection, evaluation metrics, regularization, class imbalance, and distribution shift
• BS in Computer Science, Statistics, Engineering, or equivalent practical experience; MS a plus
• Bonus: experience with LLM fine-tuning or RAG systems, real-time feature stores (Feast, Tecton), Spark/Flink streaming, or MLOps tooling (MLflow, Kubeflow, CxNjtbz)