Job
- Level
- Experienced
- Location
- Dresden
- Working Model
- Onsite
- Job Field
- IT, Software, DevOps
- Employment Type
- Full Time
- Contract Type
- Permanent employment
Job Summary
In this role, you will focus on the infrastructure and operation of simulation systems, integrating specific workloads into Kubernetes while developing the GPU infrastructure and CI/CD tools for seamless operations.
Job Technologies
Your role in the team
- You will join the Simulation team and focus on the infrastructure and operational foundations of these systems. You will not be building them in isolation: Wandelbots has an established infrastructure team responsible for shared infrastructure and platform capabilities. You will work closely with that team while owning the simulation-specific workloads, deployment patterns, and operational requirements.
- This is a hands-on engineering role with room to shape architecture and make technical decisions in collaboration with the teams involved.
- Build and operate the GPU infrastructure required by our simulation workloads, including NVIDIA GPU Operator, device plugins, driver lifecycle, and GPU sharing through MIG or time slicing.
- Work with the infrastructure team to integrate simulation workloads into our shared Kubernetes and cloud infrastructure.
- Make deployments declarative and reproducible using Infrastructure as Code and GitOps, and help consolidate the tooling used across today's environments.
- Own CI/CD and the container and image lifecycle for CUDA- and NGC-based workloads.
- Design scheduling and autoscaling for two distinct workload profiles: latency-sensitive interactive streaming sessions and compute-intensive Gaussian Splatting jobs.
- Turn the Gaussian Splatting pipeline into a reproducible workflow, from data capture and training to OpenUSD assets.
- Operate and extend our Omniverse and Isaac Sim streaming platform, including Kit App Streaming, session lifecycle, and tenant isolation.
- Establish the platform as an internal standard through self-service workflows, golden paths, templates, onboarding, and documentation.
- Prepare the solution for operation by partners and customers through packaged deployments, versioned releases, upgrade paths, and actionable diagnostics.
- Support deployments across cloud, on-premises, and partner-managed environments.
- Build meaningful observability using metrics, logs, GPU telemetry, and actionable alerting.
- Define the simulation-specific security and access model, including ingress, TURN and STUN for WebRTC, RBAC, secrets management, SSO, and tenant boundaries.
- Work closely with our robotics, product, and platform teams to ensure that the solution supports real development and customer workflows.
This text has been machine translated. Show original
Our expectations of you
Qualifications
- We are looking for an engineer who is interested in how complex systems are built, deployed, and operated, not only in delivering the next application feature. You should bring:
- Solid knowledge of Linux, containers, and networking.
- Good Python skills for automation, services, and data-processing workflows.
- The ability to turn working technology into a system that other teams can use independently.
- A strong sense of operational ownership, including documentation, diagnostics, maintainability, and reproducibility.
- Sehr gutes Englisch; Deutschkenntnisse sind von Vorteil.
Experience
- Experience building or operating production systems in platform engineering, infrastructure, SRE, DevOps, or systems software.
- Practical Kubernetes experience, including resource management, workload scheduling, and debugging distributed systems under load.
- Experience with Infrastructure as Code, GitOps, and CI/CD, together with the ability to evaluate tools based on the problem rather than a fixed preference.
- Experience with GPU workloads, or a strong technical foundation and clear interest in working with drivers, CUDA, and GPU scheduling.
- We do not expect candidates to have prior experience with every technology in this description. A strong systems foundation, the ability to learn unfamiliar components, and an interest in this problem space are more important than matching every item. The scope and responsibilities of the role will grow with your experience.
This text has been machine translated. Show original
What we offer
- Work on an innovative, deep tech product that adds value to the robotics world and to society! Become a crucial part of something new, big and exciting where you can truly make an impact!
- Cross-functional communication is key. You'll get to know many of your 100+ colleagues. Collaborate with great people on our international teams who value knowledge sharing and unique ways of doing things!
- Work in a value-driven environment where everything may not be perfect, but we're working on it! Your feedback is required while you receive feedback and appreciation for your work!
- You can have up to 3 days of volunteer time off to help your favorite charity.
- We offer training and mental health opportunities so that you can develop both mentally and professionally.
- Enjoy our free lunch, snacks and drinks, team events, 30 days of annual vacation + additional rest during Christmas season until New Year. If your child gets sick, we assure that 100% of your usual earnings will be paid while you take care of your little one.
This text has been machine translated. Show original
Benefits
Health, Fitness & Fun
Topics You Will Work On
Job Locations
About Your Employer
Wandelbots GmbH
Wandelbots GmbH is a forward-thinking deep-tech startup from Dresden that provides a platform for a vendor-agnostic robotics operating system. The company focuses on software-based automation, enabling even non-programmers to access automation solutions. Its proprietary hardware, the TracePen, supports user-centered software engineering principles and enhances efficiency in manufacturing.
Description
- Company Type
- Startup
- Working Model
- Onsite
- Industry
- Electronics, Automatization
System Software Engineer Cloud & Simulation
- Location
- Dresden
- Working Model
- Onsite
- Diversity
- Open for all genders
