Software Engineer - Foundations

Description

The Foundations team builds the platform that the company's other engineering teams ship on, and it runs that platform too.

The work combines software engineering with SRE and infrastructure. The team develops the services behind low-latency, high-quality serving and owns the infrastructure and operations beneath them, keeping production service level objectives (SLOs) on track.

The aim is to let product teams ship and scale without having to handle infrastructure, reliability or operations themselves.

The company protects over 100,000 organizations worldwide from cyber and AI-driven threats. Its prevention-first approach covers hybrid networks, cloud environments, digital workspaces and AI systems.

You would work with curious, capable colleagues who take ownership and question assumptions, with room to grow and contribute.

Responsibilities

• Build the latency-critical request path: gateway, request scheduling and orchestration.
• Take part in the on-call rotation and help make incident handling better.
• Tackle capacity and multi-tenancy problems on shared GPU infrastructure, covering isolation, admission control and graceful degradation.
• Run the data systems (ClickHouse, Postgres, Kafka), including their capacity, schemas and failure modes.
• Write design documents, review architecture across teams and help set engineering standards.
• Extend the current software stack and algorithms so they handle the most demanding use cases.
• Help maintain the DevOps stack and refine how the team develops software.

Requirements

• Have built and run production services in Go, Python or C++.
• Have profiled, load-tested and tuned a system where latency is critical.
• Work independently, write clearly and take ownership of unclear problems.
• Keep the customer behind the metric in view.
• Are curious about AI security.
• Are pragmatic and hands-on, getting things done while keeping the long-term vision in mind.
• Have experience serving models on GPUs.
• Have operated at least one stateful data system in production.
• Have experience running services across more than one cloud.
• Have contributed to open-source projects.