Senior Site Reliability Engineer at Photoroom
Paris, France, France
Posted
Remote · full_time · Visa sponsorship
Apply for this role at Photoroom
Responsibilities:
- Own and evolve ML inference infrastructure running on GPU fleets to serve millions of requests daily.
- Design cloud-agnostic, cost-efficient deployment patterns and orchestration for latency-sensitive and large-payload workloads.
- Build and tune systems for load balancing, autoscaling, queuing, and workload orchestration to maintain consistent performance.
- Implement monitoring, alerting, and observability (e.g., Datadog); analyse metrics to drive optimisations and capacity planning.
- Collaborate with ML, Product, Web, and Mobile teams to identify bottlenecks and improve deployment/workflow processes.
- Participate in incident response, runbook creation, and optionally on-call rotations to maintain reliability.
Requirements:
- Proven experience designing and operating large-scale distributed systems with high availability requirements.
- Hands-on experience with containerized workloads (Kubernetes), GPU-based inference, load balancing, autoscaling, and queuing systems.
- Strong skills in monitoring, incident management, performance tuning, and cost optimisation for ML workloads.
- Ability to make pragmatic technical decisions, communicate trade-offs, and work effectively across cross-functional teams.
- Comfortable in a fast-growing startup environment and able to take ownership of end-to-end infrastructure projects.