Key Notes
Production machine-learning experience and engineering leadership are requested for this London role building a new MLOps team. You will own deployment and operation across batch and online models, including LLMs and agents. Python and cloud infrastructure feature in the requirements. Office attendance is at least 60% over twelve weeks.
What You'll Work On
- Build the MLOps engineering team and define shared tooling, infrastructure and delivery practices for Trainline’s production machine-learning and AI systems.
- Manage model deployment and operation across batch and online services, applying testing, monitoring and observability with engineering and data-science colleagues.
- Develop maintainable cloud and CI/CD approaches, coordinating technology choices and standards across product, software, data and machine-learning delivery teams.
Why This Role Matters
- Trainline’s customer-facing ML features depend on observable production models, with deployment controls and monitoring addressing the transition from experimentation to live service.
- Shared deployment tools and operating practices let Trainline teams support different production model types through common infrastructure, reducing the need to recreate monitoring and release processes.
What They Are Looking For
- Experience: Relevant experience includes engineering leadership/mentoring and model productionisation at scale.
- Technical: Relevant technical capabilities include Python, cloud/DevOps and MLOps tooling, alongside feature-store understanding.
- Preferred: Preferred attributes include AWS, Spark/PySpark and both batch and online ML experience.