{"id":2581,"date":"2026-07-08T20:42:25","date_gmt":"2026-07-08T20:42:25","guid":{"rendered":"https:\/\/coderseditor.com\/itjobs\/job\/staff-software-engineer-inference\/"},"modified":"2026-07-08T20:42:25","modified_gmt":"2026-07-08T20:42:25","slug":"staff-software-engineer-inference","status":"publish","type":"job_listing","link":"https:\/\/coderseditor.com\/itjobs\/job\/staff-software-engineer-inference\/","title":{"rendered":"Staff Software Engineer, Inference"},"content":{"rendered":"

Staff Software Engineer, Inference Overview<\/h2>\n\n\n\n\n\n\n\n\n
Company Name<\/strong><\/td>\nAnthropic<\/td>\n<\/tr>\n
Job Role<\/strong><\/td>\nStaff Software Engineer, Inference<\/td>\n<\/tr>\n
Qualifications<\/strong><\/td>\nNot Specified<\/td>\n<\/tr>\n
Category<\/strong><\/td>\nIT Jobs<\/td>\n<\/tr>\n
Job Type<\/strong><\/td>\nFull Time<\/td>\n<\/tr>\n
Location<\/strong><\/td>\nLondon<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

This role sits on the inference engineering team responsible for the systems that power Claude for millions of people around the world. The work spans the full serving stack, from routing incoming requests intelligently to coordinating large fleets of compute across a wide range of AI accelerators. The team\u00e2\u0080\u0099s mission is twofold: keep serving costs and compute usage efficient enough to support rapid customer growth, while also building the high-performance infrastructure needed for cutting-edge research and the next generation of models.<\/p>\n

The environment is technically demanding and highly distributed. You would be working on infrastructure that runs across multiple accelerator families, emerging AI hardware, and several cloud platforms. The role is aimed at someone who can operate end to end, identify the biggest blockers in the serving path, and help remove them so production systems remain fast, reliable, and scalable.<\/p>\n

What you would do<\/h3>\n
    \n
  • Own and improve the infrastructure that serves Claude at very large scale, including the systems that receive requests, route them intelligently, and orchestrate compute across the fleet.<\/li>\n
  • Track down and eliminate infrastructure bottlenecks that affect both customer-facing inference and the internal research workflows that depend on the same serving platform.<\/li>\n
  • Design routing approaches that distribute traffic efficiently across thousands of accelerators and adapt to changing demand patterns.<\/li>\n
  • Help the organization scale compute capacity up and down dynamically so production traffic, research jobs, and experimental workloads are all supported effectively.<\/li>\n
  • Build and maintain deployment pipelines that can safely roll out new models to a global user base.<\/li>\n
  • Work on integrating new accelerator hardware and cloud environments so the platform remains flexible and not tied to a single vendor or architecture.<\/li>\n
  • Contribute to new serving capabilities, including features such as structured sampling and prompt caching, and help adapt the platform for new model architectures.<\/li>\n
  • Use observability and production data to understand how the system behaves in practice and tune performance based on real-world workloads.<\/li>\n
  • Support global serving needs, including multi-region deployments and geographic request routing.<\/li>\n
  • Collaborate on difficult distributed systems problems involving load balancing, traffic management, batching, caching, orchestration, and reliability.<\/li>\n<\/ul>\n

    What the team is looking for<\/h3>\n
      \n
    • Significant software engineering experience, especially in distributed systems or similarly complex infrastructure.<\/li>\n
    • Practical experience with performance optimization, large-scale service orchestration, intelligent routing, or related systems work.<\/li>\n
    • Exposure to machine learning systems at scale is valuable, particularly if you have worked on deployment or serving infrastructure.<\/li>\n
    • Knowledge of LLM inference optimization, batching approaches, and caching strategies is strongly preferred, though not mandatory.<\/li>\n
    • Experience with Kubernetes and cloud platforms such as AWS or GCP would be helpful.<\/li>\n
    • Programming experience in Python or Rust is desirable.<\/li>\n
    • A bachelor\u00e2\u0080\u0099s degree, or an equivalent combination of education, training, and experience.<\/li>\n
    • Relevant study or professional background in a field connected to the role.<\/li>\n
    • The seniority for this position is tied to the company\u00e2\u0080\u0099s internal level expectations rather than a fixed number of years, so the exact experience requirement is not specified.<\/li>\n
    • A practical, impact-focused mindset, with comfort stepping in wherever needed even if the work falls outside a narrow job description.<\/li>\n
    • Curiosity about machine learning systems and infrastructure, plus a desire to deepen that expertise.<\/li>\n
    • Comfort working in a setting where technical quality has direct consequences for both business performance and research outcomes.<\/li>\n
    • Strong communication skills and the ability to work well in a collaborative, discussion-heavy team.<\/li>\n
    • Awareness of the broader social and ethical implications of AI and a genuine interest in those issues.<\/li>\n
    • The company explicitly notes that applicants should not self-reject if they do not match every item on the list.<\/li>\n<\/ul>\n

      Examples of work across the organization<\/h3>\n
        \n
      • Creating routing algorithms that decide how requests should be spread across thousands of accelerators.<\/li>\n
      • Automatically scaling the compute fleet so supply matches demand across production, research, and experimental use cases.<\/li>\n
      • Building deployment systems that can release new models safely to millions of users.<\/li>\n
      • Adding support for new accelerator platforms to preserve the company\u00e2\u0080\u0099s hardware-agnostic approach.<\/li>\n
      • Developing new inference features such as structured sampling and prompt caching.<\/li>\n
      • Supporting serving for new model architectures as they are introduced.<\/li>\n
      • Analyzing observability signals to improve performance based on how the system behaves in production.<\/li>\n
      • Running multi-region deployments and geographic routing for customers around the world.<\/li>\n<\/ul>\n

        Compensation and working setup<\/h3>\n

        The annual salary range for this position is \u00c2\u00a3325,000 to \u00c2\u00a3390,000 GBP. The company states that staff are expected to work from one of its offices at least 25% of the time under its location-based hybrid policy, and some positions may require a greater office presence.<\/p>\n

        In addition to pay, the company offers optional equity donation matching, generous vacation time, generous parental leave, flexible working hours, and a collaborative office space. The role is based in London, UK.<\/p>\n

        Additional information<\/h3>\n

        Applications are reviewed on a rolling basis and there is no fixed deadline. The company also highlights that it is a public benefit corporation headquartered in San Francisco and describes its culture as highly collaborative, research-driven, and focused on high-impact work. It emphasizes communication, frequent research discussions, and a preference for work that advances long-term goals around steerable and trustworthy AI.<\/p>\n

        The posting also includes guidance about application safety: recruiters should only contact candidates from Anthropic email addresses, and applicants should be cautious of messages from other domains or requests for money, fees, or banking details before the first day. If there is any doubt about a communication, candidates are directed to use the company careers site directly.<\/p>\n

        Applicants are asked to complete the online application form, which requests standard contact details, a resume or LinkedIn profile, and responses to questions about availability, AI-use policy acknowledgement, motivation for joining the company, visa sponsorship needs, relocation, and work location. The form also asks whether the candidate is open to working in person in the office at least 25% of the time.<\/p>\n

        The company says it does sponsor visas. It explains that sponsorship is not guaranteed for every role or every applicant, but that if an offer is extended it will make every reasonable effort to secure the necessary visa and will involve an immigration lawyer to help with the process.<\/p>\n\n

        Degree Requirement:<\/strong> Not Specified<\/p>\n

        Visa Sponsorship Promising<\/strong><\/p>\n","protected":false},"author":1,"featured_media":0,"template":"","meta":{"_acf_changed":false,"om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"_promoted":"","_job_location":"London","_application":"https:\/\/job-boards.greenhouse.io\/anthropic\/jobs\/5097742008","_company_name":"Anthropic","_company_website":"","_company_tagline":"","_company_twitter":"","_company_video":"","_filled":0,"_featured":0,"_remote_position":0,"_job_salary":"\u00a3325000 - \u00a3390000 per yearly","_job_salary_currency":"GBP","_job_salary_unit":"yearly"},"job-types":[38],"class_list":["post-2581","job_listing","type-job_listing","status-publish","job-type-experienced"],"acf":[],"jetpack_sharing_enabled":true,"jetpack_likes_enabled":true,"_links":{"self":[{"href":"https:\/\/coderseditor.com\/itjobs\/wp-json\/wp\/v2\/job-listings\/2581","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/coderseditor.com\/itjobs\/wp-json\/wp\/v2\/job-listings"}],"about":[{"href":"https:\/\/coderseditor.com\/itjobs\/wp-json\/wp\/v2\/types\/job_listing"}],"author":[{"embeddable":true,"href":"https:\/\/coderseditor.com\/itjobs\/wp-json\/wp\/v2\/users\/1"}],"wp:attachment":[{"href":"https:\/\/coderseditor.com\/itjobs\/wp-json\/wp\/v2\/media?parent=2581"}],"wp:term":[{"taxonomy":"job_listing_type","embeddable":true,"href":"https:\/\/coderseditor.com\/itjobs\/wp-json\/wp\/v2\/job-types?post=2581"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}