Staff Machine Learning Engineer

This role builds, runs and scales the production inference platform behind search and personalization for the global streaming app Max and future streaming initiatives. Every recommendation, search result and personalised row is served by a model on this platform, which must answer in milliseconds for 100+ million users worldwide. The ML Infrastructure & Tools (MLIT) team owns the deployment, lifecycle and inference of models for the Search & Personalization organisation, operating a shared self-service inference platform on Kubernetes across multiple frameworks and runtimes behind one common control plane.

• Own the architecture and roadmap of the model serving runtime, supporting multiple ML frameworks and inference engines behind one control plane.
• Lead large-scale migrations and rollouts of production models onto the platform with zero viewer-facing regressions.
• Build and maintain serving backends, model configurations and batching strategies that meet strict tail-latency targets at high request volume.
• Design autoscaling, capacity management and multi-region resilience for latency-critical serving across Max, Discovery+ and other WBD offerings.
• Drive observability, cost optimisation and performance tuning across the serving fleet, including hardware selection, GPU and CPU utilisation, and model optimisation.
• Define the contract between model training and serving so validated model artifacts deploy to production through standard automated paths, and establish per-model SLOs with model owners and telemetry that proves them.
• Collaborate with applied scientists, partner engineering teams and the core infrastructure team as the platform onboards new models, runtimes and markets.
• Author, test, review and optimise production-level code in Python while applying best practices in version control, CI/CD and code integration.
• Take part in the platform on-call rotation and lead incident reviews that result in design changes, not just fixes.
• Contribute design documents and technical documentation, mentor engineers, raise the bar in design and code reviews, and represent the platform in cross-team architecture decisions.

• 9+ years of industry experience.
• 5+ years of experience building or operating production ML infrastructure or comparable low-latency, large-scale distributed systems with a large customer base.
• Hands-on experience running model inference in production at scale.
• Strong Kubernetes fundamentals and experience running GPU and CPU workloads on it.
• Proven performance engineering skills: profiling, tail-latency analysis, batching and concurrency under load.
• 7+ years of programming experience in Python, plus working fluency in a systems language such as C++, Go or Rust.
• Good practical knowledge of the modern machine learning lifecycle, from training artifacts through deployment, rollout and monitoring.
• Strong architectural and design skills, with a track record of leading multi-quarter technical initiatives across teams.
• Operational maturity: experience owning a tier-1 service, including on-call.
• Experience with one of the cloud platforms AWS/GCP/Azure, Docker and Kubernetes, and CI/CD tools like GitHub Actions, Jenkins etc.
• Excellent written and verbal communication skills; comfortable contributing to technical documentation and on-call support.
• Ability to technically collaborate with partners across time zones.

• Experience extending an inference server with custom backends, or contributing to an open-source serving runtime.
• Background in recommendation, ranking or search systems and their latency and freshness constraints.
• Experience building an internal platform that other teams adopted self-service.
• Working experience with global teams.