ML Platform Engineer

About Agile Robots SE and the Role

Agile Robots SE, a cutting-edge high-tech company headquartered in Munich, Germany, specializes in bridging artificial intelligence and robotics by integrating advanced force-moment sensing with top-tier image processing technologies. With over 2300 employees worldwide, Agile Robots develops intelligent, user-friendly, and cost-efficient robotic solutions focused on precision assembly. The AI Research and Development Group seeks a skilled ML Platform Engineer to design, implement, and manage scalable infrastructures that enable seamless transition of machine learning models from prototype phases to robust production deployments.

Key Responsibilities

• Design and scale distributed training workflows for large-scale models utilizing frameworks like PyTorch Distributed and DeepSpeed, alongside cluster schedulers such as SLURM or Kubernetes.
• Develop and maintain containerized machine learning environments that facilitate reproducibility in experimentation and benchmarking.
• Create and manage CI/CD pipelines tailored for machine learning systems to ensure reliable testing, training, and deployment of models.
• Implement lifecycle management solutions including experiment tracking, model versioning, and reproducibility workflows using tools like ClearML or Weights & Biases.
• Establish monitoring and observability systems such as Prometheus and Grafana to oversee model performance and system health, and to detect production drift.
• Collaborate cross-functionally with research, data, and robotics teams to integrate new models into stable production systems.

Essential Qualifications and Skills

• A degree in Computer Science, Software Engineering, or a related domain, complemented by professional experience in developing and maintaining machine learning or software infrastructures in production environments.
• Proficiency in distributed training architectures operated on Kubernetes and Docker platforms using PyTorch Distributed, DeepSpeed, and scheduling tools like SLURM.
• Experience designing and executing CI/CD pipelines supporting machine learning workflows for dependable model testing, training, and deployment.
• Hands-on experience managing ML workloads on cloud infrastructure, preferably with Amazon Web Services (AWS).
• Familiarity with experiment tracking and model versioning platforms such as MLflow or Weights & Biases.
• Capability in deploying monitoring solutions and drift detection mechanisms leveraging Prometheus and Grafana.
• Strong software engineering background with expertise in Python and system design associated with ML systems beyond early prototypes.

Desirable Competencies

• Experience working on large-scale or multimodal machine learning systems, including vision-language-action models.
• Knowledge of infrastructure-as-code tools, like Terraform.
• Exposure to ML pipeline orchestration and management tools.
• Familiarity with high-performance or distributed computing environments.

What the Company Offers

• A dynamic and financially stable high-tech environment supported by world-class investors.
• An international and interdisciplinary team exceeding 60 nationalities working collaboratively.
• Extensive opportunities for professional growth fueled by the company's ongoing expansion.
• Impactful projects and challenging assignments alongside experts fostering both personal and career development.
• Corporate benefits including health, mobility, and learning programs with a monthly net allowance of 100 €.
• State-of-the-art office facilities featuring a rooftop terrace overlooking Munich, complimentary drinks and fruit, and regular company events enhancing workplace experience.

Diversity and Inclusion

Agile Robots SE values diversity and inclusion, encouraging applications regardless of gender identity, nationality, ethnicity, religion, age, disability, or sexual orientation.