DevOps Engineer

About Bones Bones combines advanced technology with deep knowledge of human movement. We have worked on AAA games, film projects and motion research. Our flagship project is one of the largest human motion datasets, which we license to major technology companies to train digital humans and robots.

Where we are Our on-premise infrastructure is up and running, with hardware handled by our IT team. Our tools developers and ML engineers build the processing pipelines that turn raw capture into client-ready datasets. Your job is to own the platform those pipelines run on. Our data is growing fast: more capture, more processing, more ML. You’ll keep the platform reliable, help the team run their pipelines smoothly at scale, and reshape the infrastructure behind it as our needs grow.

What you will do

• Own the infrastructure our processing and ML pipelines run on, and be the go-to person when something breaks.
• Support our tools developers and ML engineers in running, scaling and debugging their pipelines.
• Evolve the infrastructure toward a properly distributed setup: job scheduling across our CPU and GPU machines, queues, retries and better resource sharing.
• Containerize processing jobs and make them easy to run reliably, including on GPU machines.
• Automate the configuration of our servers and environments as code.
• Build and maintain CI/CD for our internal tools and ML code.
• Set up monitoring and alerting so failed jobs, full storage and idle machines are caught early.
• Keep our compute and storage well used, and work with IT on capacity planning as data volumes grow.
• Spot infrastructure bottlenecks, propose improvements and drive them through.

What we are looking for

• 2+ years of hands-on DevOps, SRE or platform engineering experience.
• Experience owning production infrastructure, not just contributing to it.
• Strong Linux skills and scripting in Python and Bash.
• Docker and CI/CD (GitHub Actions, GitLab CI or similar).
• Experience running batch or distributed workloads (Kubernetes, Slurm, Nomad, Ray or similar).
• Configuration management or infrastructure as code (Ansible, Terraform or similar).
• Experience with large-scale storage (NAS, Ceph, MinIO or similar).
• Monitoring and logging (Prometheus, Grafana or similar).
• An efficiency-minded approach: you notice idle machines and wasted storage and fix them.
• A supportive attitude: you enjoy making other engineers more productive.
• Comfort with ownership: you take responsibility when things break and push improvements without waiting to be asked.
• Comfort working with AI coding tools.
• Communicative English.

Nice to have

• GPU workloads and ML infrastructure (CUDA, model training or inference jobs).
• Moving and storing large datasets (terabytes of video and 3D data).
• Familiarity with pipeline orchestration tools (Airflow, Dagster, Prefect or similar).
• Some cloud experience (AWS, GCP or Azure), in case we extend into hybrid later.
• Interest in motion capture, 3D, robotics or generative AI.

What we offer

• Ownership of the infrastructure behind an ambitious project with major technology companies,
• Work with experienced motion specialists and ML engineers on projects for major technology companies.
• Hybrid work from our studio at Strażacka 53, Warsaw.
• Employment contract (umowa o pracę), or another form if you prefer.
• Quarterly bonuses and flexible working hours.
• Conferences and an individual development budget.
• Private healthcare and a Multisport card.
• Excellent coffee