MLOps Engineer

We're working with a London scale-up building proprietary AI that operates under strict real-time performance constraints. Their technology is deployed in production environments where reliability and latency aren't negotiable. If you want something more exciting and higher stakes than a standard SaaS ML problem, this is for you.

This is a genuine engineering role with real ownership, not a support function for researchers.

The Role

You'll own the infrastructure that moves models from research into production and keeps them performing reliably at scale. You'll work closely with research and engineering teams, sitting at the boundary between experimentation and production systems.

What you'll be doing:

• Designing and maintaining end-to-end ML pipelines across the full lifecycle
• Deploying and monitoring models in production across multiple compute environments
• Building observability and alerting layers to catch performance regressions before they reach users
• Implementing CI/CD pipelines with automated quality checks
• Enabling reproducible experimentation for the research team
• Managing versioning across datasets, features, and models

What we're looking for:

• Strong Python and software engineering fundamentals
• GCP experience is a real bonus
• Experience with low-latency or real-time systems
• Startup background - comfortable in a small, fast-moving team
• Domain-specific ML experience (signal processing, audio, speech) is a genuine advantage

The Package:

• Share options
• Private medical insurance
• ~38 days holiday
• Hybrid - 3 days in office per week, central London