Senior Site Reliability Engineer

The role

You will plan and carry out the work needed to grow Encord's platform from hundreds to thousands of AI teams and from terabytes to petabytes of data. You will embed in teams that build and run core infrastructure, making the platform performant, reliable, observable and scalable.

What you will do

• Profile and optimise services handling large-scale data pipelines and plan capacity for storage and compute.
• Work with squads to set performance benchmarks.
• Partner with backend and ML engineers to improve CI/CD pipelines and review infrastructure changes.
• Define and own SLIs, SLOs and SLAs for critical services.
• Build alerting, runbooks and incident response; lead blameless postmortems.
• Design, deploy and maintain cloud infrastructure on GCP and AWS, including Kubernetes clusters, networking and petabyte-scale storage.
• Improve developer productivity and guide automation and tooling across engineering.
• Instrument services with distributed tracing, logging and metrics so every service is observable before production.

Required qualifications

• Hands-on experience in SRE, DevOps, platform engineering or similar in a production environment.
• Strong fundamentals in designing, building and maintaining resilient distributed and/or high performance systems.
• Solid understanding of networking, operating systems and database technologies.
• Experience with observability fundamentals: metrics, logs, traces and alerting.
• Comfortable with on-call rotations and incident management.
• Open to learning the tech stack: Python, Rust, TypeScript, React, Kubernetes and GCP.

Benefits

• Competitive salary, commission and meaningful equity.
• 25 days annual leave plus UK public holidays.
• Annual learning and development budget.
• Daily company lunches.
• Monthly socials and bi-annual team offsites.
• Travel for customer visits, events and conferences.

Before you apply

• The posting asks for an open attitude to learning: no need to have experience across the whole tech stack.

About the company

Encord is the universal data layer for AI, helping 300+ AI teams train and run models on the right data. The company has 100+ staff and has raised $60M in Series C funding.