Software Engineer (Inference Platform), London

Your Impact

We are building the largest foundation models in biotech and applying them immediately to cure disease. You will play a pivotal role in ensuring the reliability and scalability of the foundations that make this possible.

As an Engineer, you will get involved with the efforts to harden our systems, ensuring our groundbreaking AI is built on an unshakeable base, working closely with the research team and the Applied ML teams to ensure the infrastructure is stable, reliable and can operate with more data and larger models as we grow. 

What You Will Do

• Develop and operate our inference platform, serving fleets of cutting-edge machine learning models to scientific applications.
• Optimize our existing inference services. You will solve core scaling limits, ensuring high-throughput performance and feature parity across our model serving stack.
• Contribute to core technical decisions on tooling and architectural design.
• Deliver high-quality and well-tested user-focused features.

Skills and Qualifications

Essential:

• Hands-on experience deploying and scaling inference frameworks (e.g., KServe, Seldon) within Kubernetes, including a strong understanding of cloud-native ML lifecycle management.
• Strong programming skills and a "reliability-first" approach to software development.
• Experience developing, operating, and debugging large-scale distributed systems, with a strong grasp of high-throughput architectures.
• Understanding of modern infrastructure and DevOps practices (Infrastructure as Code, CI/CD pipelines, and comprehensive observability).

Nice to Have:

• Experience with Google Cloud Platform (GCP).
• Familiarity with workload scheduling, ML efficiency research, and hardware benchmarking.
• Familiarity with GPU-aware infrastructure and specialized inference servers (e.g. Triton, TF Serving).