DevOps Engineer
Description
At ThetaRay, our purpose is to make the world a safer place by protecting the integrity of the global financial system.
We do this by putting AI at the core of both our technology and our way of working. Our AI-driven solutions help banks and fintech companies worldwide detect and stop serious financial crime, from human trafficking and terrorist financing to sophisticated money laundering, while advanced technology, automation, and AI-driven tools help our teams collaborate smarter, move faster, and continuously improve how we build, deliver, and innovate.
About the role:
We're looking for a hands-on DevOps engineer to design, automate, and operate the environments our platform runs in- public cloud alongside on-premises, including restricted and air-gapped customer sites, and to build them the way infrastructure gets built in 2026: as code, with AI agents doing much of the heavy lifting and your judgment deciding what actually ships.
Two things sit on your plate. The first is our R&D team: the pipelines, paved roads, and self-service tooling that let engineers ship several times a day without asking permission or holding their breath- increasingly built and maintained alongside AI agents rather than by hand. The second is where our software actually lands: customer environments that are rarely identical, often heavily locked down, and occasionally have no route to the internet at all. Making a modern, AI-powered platform run reliably inside those constraints is the genuinely interesting part of this job.
Why Join Us
- Innovation: Modern stack, real cloud-native and GitOps practice, and AI tooling in daily use rather than in a slide.
- Autonomy: Senior engineering ownership without management overhead. Your ideas land.
- Leverage: Every environment you standardize makes the next one faster- and every workflow you hand to an AI agent stops costing you a day. You'll see your own work compound.
- Do you like to fight crime? Our software helps catch financial criminals. That's the actual output of your infrastructure work.
Responsibilities:
- Developer enablement: Continuously improve the tools, pipelines, and processes our R&D team depends on- including how AI coding agents plug into them safely. If something is slow, manual, or superstitious, you're empowered to fix it.
- Infrastructure and automation: Design, implement, and automate our cloud infrastructure and on-premises OpenShift environments as code, not as runbooks- including the compute and GPU capacity our AI workloads run on.
- Deployment and delivery: Build declarative delivery with Argo, Helm/Helmfile, and Kubernetes, so a new environment is a configuration change rather than a bespoke effort. This includes getting our AI-driven platform into restricted and disconnected sites.
- AI-assisted engineering: Use AI agents as a normal part of how automation gets written, reviewed, and debugged- and help set the standards that keep agent-written infrastructure safe to merge: scoped permissions, policy checks, and human review where it counts.
- On-call and escalation: Join the on-call rotation, support lower tiers through incidents, and push root-cause fixes over repeat mitigations.
Requirements
- Experience: 5+ years in DevOps, Platform Engineering, or SRE, including ownership of production systems.
- Debugging: Strong, systematic troubleshooting under production pressure - across Linux, networking, and Kubernetes.
- Kubernetes: Real experience in both development and production environments (OpenShift an advantage).
- CI/CD: Hands-on pipeline authoring (Jenkins an advantage - it's a meaningful part of our current stack).
- Infrastructure as Code: IaC pipelines in practice (Terraform an advantage).
- Cloud: A major cloud provider (Azure a plus).
- Delivery tooling: Argo, Helm, Helmfile, or equivalent declarative deployment tooling.
- Linux and scripting: Proficient in Linux, with Python or Bash.
- Observability: Prometheus, Grafana, ELK, or similar.
- AI-assisted workflow: Comfortable working with AI coding agents day to day- and clear-eyed about where they need supervision.
- FinOps: Practical experience keeping cloud spend deliberate.
Nice to Have
- Experience running AI/ML or inference workloads (GPU scheduling, vLLM, KServe, Ray, vector databases).
- Genuine appetite for new technology and continuous learning.
- Strong judgment about production risk- you know which changes deserve caution.
- Good with people. You'll work across R&D, support, and occasionally customer-side engineers.