About Us

DeepJudge is Switzerland’s leading AI and ICT scale-up, transforming how law firms and legal departments access and leverage their knowledge. Founded by former Google search engineers with PhDs in AI from ETH Zurich, and backed by top-tier Silicon Valley investors, we are building the intelligence layer that powers the legal industry.

Every firm can license the same AI models, but no two firms share the same institutional knowledge. Decades of experience, work product, and precedent are often trapped across systems, underused and inaccessible. DeepJudge makes this expertise instantly available through world-class enterprise search and AI infrastructure, enabling firms to automate workflows, build knowledge-powered applications, and turn experience into a lasting advantage.

Our technology is trusted by many of the world’s leading law firms, including Freshfields, Gunderson Dettmer, Holland & Knight, Arent Fox, and Cozen O’Connor. Headquartered in Switzerland with a growing team across North America, we are expanding rapidly and shaping the future of how professional knowledge is discovered and applied.

At DeepJudge, we move fast, think rigorously, and care deeply about what we build. We combine cutting-edge research with close collaboration with our clients to deliver solutions that truly make an impact. If you want to be part of a team defining how AI transforms high-stakes knowledge work, this is the place to do it.

About the role

As a Site Reliability Engineer at DeepJudge, you own the reliability of the systems that leading law firms depend on every day. We run our AI search platform in dedicated production environments, each a large-scale search and retrieval system operating at very high data volumes, alongside our ingestion pipelines and data stores, on container-orchestrated infrastructure across multiple public and private clouds. Keeping those environments fast, available, and trustworthy is a genuine engineering challenge, and it's your mission.

This is a technical, infrastructure-focused role: you'll spend your time in systems, code, and telemetry. You'll harden the platform so that classes of failure can't recur, build observability and SLOs to let us see problems before they surface, plan capacity ahead of relentless data growth, and automate away the toil of safely operating many production environments. You'll work across our deployment and infrastructure-as-code and partner closely with product engineering to feed reliability learnings back into the platform. If you like operating serious distributed systems at scale and turning firefighting into engineered reliability, this role is for you.

Your responsibilities will include

Owning the availability, latency and performance of our production environments: large-scale search and retrieval services, ingestion and data stores on container-orchestrated infrastructure across multiple public and private clouds

Owning the cloud infrastructure across multiple public and private clouds, networking, identity and access, compute and storage, and managing cloud capacity, quotas and cost as the fleet grows

Improving deployment and infrastructure through code: GitOps-based configuration and deployment and infrastructure-as-code across our public and private clouds, with safe, previewed, recoverable production changes

Defining and instrumenting SLOs and error budgets and building the dashboards, metrics and alerting that make production observable and actionable

Reducing operational toil through automation and internal tooling written in Python and / or Go

Provisioning and standardizing new production environments to a defined reliability bar

You're a great fit if you

Have experience in SRE, production / platform engineering, or infrastructure-heavy software engineering, operating real production systems under reliability expectations

Are strong at container-orchestration operations in production and comfortable on at least one major public cloud

Have solid cloud infrastructure experience on a major public cloud and a working grasp of cloud capacity, quotas, and cost