GenAI Systems Engineer

Position Summary :

Knorr-Bremse, the world leader in braking systems and safety-critical mobility solutions, is building an AI Center of Excellence to accelerate innovation across truck and rail businesses.

We are seeking a skilled ML / GenAI Engineer to be the operational backbone of our AI team. Your mission is to take the innovative models and prototypes developed by our scientists and transform them into production-grade, scalable, and highly efficient services.

This is a hands-on engineering role focused on the post-modeling lifecycle : optimization, deployment, and MLOps. You will be the expert on making our cutting-edge Generative AI systems (including LLMs and RAG pipelines) run fast, consume minimal resources, and operate reliably at scale. You will bridge the critical gap between experimental AI and robust, enterprise-ready applications.

If you are passionate about the engineering challenges of making large models performant and cost-effective in real-world environments, this role is for you.

Essential Functions :

• Model Optimization and Inference Acceleration : Analyze and optimize inference scripts for low latency and high throughput. Implement advanced techniques to reduce model memory consumption and computational cost, such as quantization (e.G., bitsandbytes, GPTQ), pruning, and knowledge distillation. Leverage specialized serving runtimes and compilers (e.G., vLLM, TensorRT-LLM, Triton Inference Server) to maximize hardware utilization.
• Deployment and Infrastructure Management : Containerize AI applications and their dependencies using Docker for reproducible deployments. Deploy models as scalable and resilient microservices on cloud platforms (AWS, GCP and / or Azure) using services like Kubernetes (GKE / AKS), Cloud Run, or Vertex AI Endpoints. Build and manage the cloud infrastructure required to support our AI services, using Infrastructure as Code (IaC) principles where applicable (e.G., Terraform).
• MLOps and Automation : Design and implement robust CI / CD pipelines for automated model testing, building, and deployment. Develop comprehensive monitoring and logging solutions to track model performance, resource usage, and system health in production. Implement alerting mechanisms to proactively identify and address issues like model drift, performance degradation, or infrastructure failures.
• API and Service Development : Develop clean, well-documented, and efficient APIs (e.G., REST, gRPC) using frameworks like FastAPI or Flask to serve model predictions to downstream applications. Ensure the security, authentication, and scalability of the AI service endpoints.
• Cross-Functional Collaboration : Work closely with AI scientists, product managers, software engineers, and UX designers to align AI solutions with business needs. : Work closely with the AI Scientist to understand model architecture and dependencies, providing feedback to improve its "deployability." Partner with the lead Generative AI Scientist to ensure that the production environment can meet the performance and cost requirements of their proposed solutions.
• Ensuring Responsible & Secure AI : Implement AI solutions in line with ethical and security guidelines . The AI Engineer is expected to incorporate principles of Responsible AI – ensuring models are fair, transparent, and privacy-compliant. They must guard against risks (bias, data leakage, prompt injection attacks on agents, etc.) and follow security best practices for AI deployments. For instance, when deploying an agentic AI system that can take actions, the engineer will put guardrails in place to prevent unintended behaviors and to log decisions for auditability.
• Continuous Learning & Improvement : Stay up-to-date with the latest AI research and emerging tools, and continuously improve AI models in production. AI Engineers track new developments (new model architectures, improved frameworks like LangChain updates, etc.) and assess how these can enhance existing systems. They also gather user feedback and model performance data to iteratively refine the solutions (e.G. fine-tuning an LLM for better accuracy or reducing latency by optimizing code).

Skills :

proficiency in shell scripting (Bash).

Behavioral Competencies :

Experience & Education :

Position Requirements :

What We Offer