Sr. Software Engineer, Inference

Sr. Software Engineer, Inference Overview

This role sits on the inference engineering team responsible for the systems that power Claude for users around the world. The work is centered on building and operating the large-scale infrastructure that serves model traffic reliably, efficiently, and at very high performance. The team handles the full serving stack, from intelligent request distribution through fleet-wide orchestration, and works across multiple cloud environments, accelerator types, and deployment scenarios.

What you will do

• Design, implement, and maintain the distributed systems that deliver Claude to a global audience at massive scale.
• Create resilient systems that can adjust quickly as conditions change in production.
• Build request-routing, load-balancing, and traffic-management components that spread traffic intelligently across thousands of accelerators.
• Improve fleet efficiency by designing autoscaling and orchestration approaches for production, research, and experimental workloads.
• Develop and operate deployment pipelines that safely release new models to users.
• Provide low-latency, high-throughput inference infrastructure that supports researchers working on future model generations.
• Help integrate new accelerator platforms and extend serving support to new model architectures.
• Work on the broader serving stack, including networking, scaling, orchestration, and production reliability.
• Use production telemetry and observability data to identify bottlenecks and tune system behavior.
• Support geographically distributed and multi-region deployments for customers in different locations.

What we are looking for

• Strong coding ability in Python or Rust.
• Experience as a software engineer building and operating distributed systems in production.
• Working knowledge of containerized infrastructure, especially Kubernetes, and experience with at least one major cloud provider such as AWS, GCP, or Azure.
• A practical, impact-focused approach to engineering and a willingness to adapt as priorities change.
• Comfort stepping in to help wherever needed, even when the work falls outside a narrow job description.
• Interest in learning more about machine learning infrastructure and model-serving systems.
• Ability to do well in an environment where engineering excellence directly affects both product outcomes and research progress.
• Care and thoughtfulness about the broader social effects of AI systems.
• A bachelor’s degree, or an equivalent combination of education, training, and experience.
• Relevant academic or professional background in a field connected to the role.
• Experience expectations vary based on the internal level for the position.

Preferred background

• Substantial experience with large-scale distributed systems that must perform well under heavy load.
• Experience building and deploying machine learning systems in production at scale.
• Prior work on load balancing, request routing, or traffic-management systems.
• Knowledge of large language model inference optimization, including batching and caching strategies.
• Deep operational experience running Kubernetes and cloud infrastructure at scale.
• Familiarity with AI accelerator hardware such as GPUs, TPUs, or other emerging chips.

Example work you might do

• Design routing algorithms that intelligently distribute requests across many accelerators in different environments.
• Build autoscaling systems that match compute supply to demand across production, research, and experimental workloads.
• Create production-ready deployment pipelines for dependable model releases to millions of users.
• Contribute new inference capabilities and help support new model architectures.
• Study observability data to improve performance based on real production usage.
• Manage multi-region deployments and geographic routing for a worldwide customer base.

Team and company context

The organization’s mission is to build AI systems that are reliable, understandable, and steerable, with an emphasis on safety and broad benefit to users and society. The broader company is a fast-growing group that includes researchers, engineers, policy specialists, and business leaders working together on beneficial AI. The engineering culture values collaboration, communication, and high-impact work. The company describes its research approach as highly collaborative and focused on a small number of large-scale efforts rather than many narrow projects.

The team places strong value on communication and on work that advances long-term goals around trustworthy AI. The company also emphasizes that it sees AI research as an empirical discipline and encourages candidates to review its recent research to understand the direction of the work.

Compensation and working setup

The annual salary range for this position is £225,000 to £325,000 GBP. The role follows a location-based hybrid arrangement: employees are expected to spend at least 25% of their time in one of the company’s offices, and some positions may require more in-office time.

The company offers competitive pay and benefits, optional equity donation matching, generous vacation, generous parental leave, flexible hours, and a pleasant office space designed for collaboration.

Application process and additional notes

Applications are reviewed on a rolling basis, and there is no fixed deadline. The application form asks for standard contact and resume details, a cover letter, LinkedIn profile or resume, and responses to questions about availability, relocation, prior interviews, AI policy acknowledgment, and work authorization. Candidates are also asked whether they are open to working in person at least 25% of the time and to provide the address from which they plan to work.

The company encourages applicants not to self-reject if they do not match every listed qualification, noting that strong candidates may still be a fit even if they do not meet every item exactly. It also states that it values diverse perspectives and wants applicants from underrepresented groups to feel encouraged to apply.

For safety, the company warns that legitimate recruiters will only use email addresses from its own domain, may sometimes work through vetted agencies that identify themselves clearly, and will never ask for money, fees, or banking details before the first day. If there is any doubt about a message, candidates are directed to check the careers site directly.

Visa support is available: the company says it sponsors visas, will make every reasonable effort to secure sponsorship for an offered candidate, and uses an immigration lawyer to help with the process, while also noting that sponsorship cannot be guaranteed for every role or every applicant.

Degree Requirement: Not Specified

Visa Sponsorship Promising