Senior Software Engineer – Inference Systems (London) | Visa + Salary Insights

📍 Location: gb

🏷 Type: Full-time

Job Intelligence

📍 More in this location:
Browse AI jobs in gb

🏷 Similar roles:
View similar Full-time jobs

🌐 Explore all jobs:
View all AI job listings

Job Overview

This Senior Software Engineer (Inference Systems) role at Anthropic focuses on building large-scale distributed infrastructure powering Claude for millions of users globally. The position sits within the Inference team, responsible for request routing, model serving, and compute optimization across heterogeneous AI accelerators. The ideal candidate is a strong systems engineer with deep experience in distributed systems, high-performance infrastructure, and production-grade machine learning serving systems. This is a high-impact role where engineering decisions directly affect latency, cost efficiency, and global AI model availability across multiple cloud environments.

📅 Job Timeline & Status

This role is in an actively hiring phase with rolling review, meaning applications are evaluated continuously rather than in fixed batches. Given Anthropic’s rapid infrastructure scaling and compute expansion, urgency is high. Candidates with strong distributed systems and inference optimization experience should apply immediately, as similar roles in frontier AI labs typically receive extremely high global application volume within days of posting.

🌍 Work Eligibility & Location

This is a globally accessible role with strong visa sponsorship support for high-performing candidates. The hybrid requirement ensures deep collaboration with research and infrastructure teams, especially during system design, deployment cycles, and performance tuning across distributed inference clusters.

💰 Salary Intelligence

This is a top-tier compensation band for senior distributed systems engineers in frontier AI. The package is highly competitive within global AI infrastructure roles, especially considering equity upside and exposure to large-scale model serving systems used by millions of users.

📊 Role Breakdown

This role centers on building and scaling high-performance inference infrastructure that powers production AI systems at global scale. Engineers will design systems that handle millions of requests per second, optimizing latency, throughput, and compute efficiency across heterogeneous hardware environments including GPUs and specialized AI accelerators. Approximately 40% of the work involves distributed systems engineering, while 30% focuses on inference optimization, batching strategies, caching layers, and routing algorithms. Another 20% involves cloud orchestration across AWS and GCP, and the remaining 10% is dedicated to observability and performance tuning.

Key responsibilities include building intelligent request routing systems, optimizing model serving pipelines, and improving fleet-wide autoscaling systems. Engineers will work closely with research teams to support new model architectures and ensure production readiness. Strong proficiency in Python or Rust is required, along with deep experience in Kubernetes-based deployments. The role demands continuous analysis of real-world production traffic to improve efficiency and reduce inference cost per token while maintaining strict reliability and uptime guarantees.

🧩 Required Skills & Fit

📈 Difficulty & Competitiveness

This is a very high difficulty role due to its intersection of cutting-edge AI infrastructure and large-scale distributed systems. Candidates are expected to demonstrate production-level ownership of complex systems under real-world constraints, including latency, cost efficiency, and hardware heterogeneity. Competition is intense, with applicants typically coming from top-tier tech companies and AI labs.

🚀 Career Impact

This role offers exceptional long-term career acceleration due to direct exposure to frontier-scale AI infrastructure. Engineers gain deep expertise in production inference systems, making them highly competitive for senior engineering and leadership roles in top AI labs and hyperscale infrastructure teams.

📋 Key Responsibilities

The role involves designing and maintaining distributed inference systems that serve large language models at global scale. Engineers will build and optimize intelligent routing systems that distribute traffic across thousands of accelerators while maintaining low latency and high availability. Responsibilities include improving batching strategies, implementing caching layers for prompt reuse, and enhancing autoscaling systems that dynamically adjust compute resources.

Additional responsibilities include integrating new hardware platforms, supporting experimental model deployments, and ensuring seamless multi-region service reliability. Engineers will also work on observability pipelines, analyzing production metrics to continuously refine performance. Collaboration with research scientists is essential to ensure new model architectures are efficiently deployed into production environments.

🎯 Application Strategy

🧠 Application Optimization (Adaptive)

This section is personalized by seniority.
How to use: Paste into ChatGPT, Claude, or Gemini

🧠 Fit & Positioning Analysis

Evaluate your match before applying.

📅 Application Signals

This role shows strong application urgency due to ongoing scaling of inference infrastructure. With no fixed deadline and rolling review, applications may close without notice depending on hiring velocity. Competition is expected to be extremely high, particularly from senior engineers in AI infrastructure and distributed systems domains. Interview loops for similar roles typically move quickly once shortlisted, with initial feedback often within weeks.

🔗 Apply for this Job

Apply on Company Site

✅ Job Source & Verification

This listing is derived directly from the official Anthropic careers posting, making it a highly credible primary source. Details such as compensation, hybrid policy, and role responsibilities are consistent with publicly available company hiring standards. Last updated: 2026-06-14, ensuring the information reflects the most recent verification at time of publication.