Sr. Machine Learning Engineer

QSC; Zürich; Hybrid; CHF 130.000 - 180.000; Vollzeit

Zusammenfassung

Q-SYS VisionSuite in Zurich, hybrid, seeks a Senior ML Engineer to evaluate, integrate, and optimize state-of-the-art ML models powering perception and awareness for intelligent AV systems.

You will benchmark models, design reproducible eval pipelines, and deploy production ML components with Robotics and Platform teams, ensuring real-time performance on constrained hardware while maintaining robustness and observability.

Qualifikationen

• MSc or PhD in Computer Science, Engineering, Robotics, or related field.
• 5+ years of hands-on ML engineering experience.
• Proven experience integrating ML models into production systems.
• Strong Python skills with PyTorch, TensorFlow, or ONNX.
• Solid software engineering fundamentals including CI/CD, testing.
• Experience optimizing models for real-time or constrained environments.
• Understanding system-level trade-offs in latency-sensitive architectures.
• Ability to work independently and drive technical decisions.
• Excellent cross-functional collaboration and communication.
• Familiarity with ROS, TensorRT, or MLOps tools.

Aufgaben

• Evaluate and benchmark state-of-the-art ML models for perception, tracking, and multimodal awareness.
• Design and maintain evaluation pipelines measuring performance, latency, memory footprints, and robustness.
• Integrate ML models into production systems with Robotics and Platform teams.
• Optimize inference pipelines for real-time performance on constrained hardware (CPU/GPU/edge).
• Improve model efficiency using quantization, pruning, distillation, and runtime optimization.
• Write production-grade Python and C++ following clean architecture and modular design.
• Contribute to CI/CD pipelines, automated testing, regression validation, and monitoring for ML components.
• Ensure reproducibility, versioning, and traceability of models, datasets, and experiments.
• Collaborate to industrialize prototypes into scalable production systems.
• Work with Product and System Architects to align ML solutions with hardware roadmap constraints.

Kenntnisse

5+ years of hands-on ML engineering

Production‑ready ML

Python proficiency

CI/CD experience

Independent work / decision making

PyTorch / TensorFlow / ONNX

C++ in performance environments

ROS / TensorRT / MLOps tools

Computer vision / multimodal systems

Ausbildung

MSc or PhD in CS / Engineering / Robotics

Tools

Docker

MLflow

Weights & Biases

ONNX

Jobbeschreibung

Overview

As a Senior ML Engineer in the intelligent AV pod, you will be responsible for evaluating, integrating, and optimizing state‑of‑the‑art machine learning models that power the perception and awareness engine behind Q‑SYS VisionSuite.

This position emphasizes strong engineering execution: systematically benchmarking external and internal models, selecting the right techniques for production constraints, and ensuring robust deployment in real‑time, resource‑constrained AV environments.

You will work closely with ML, Robotics, and Software Engineers to advance VisionSuite as a reliable, maintainable, and high‑performance solution for smart meeting spaces and intelligent buildings.

This position is based in Zurich, Switzerland (hybrid).

Your mindset

• Engineering‑First ML Practitioner: You prioritize robustness, reliability, and maintainability over novelty.
• Strong Software Engineer: You design modular, testable, and extensible systems and apply software engineering best practices consistently.
• Production‑Oriented Thinker: You consider latency, memory, hardware constraints, observability, and lifecycle management from day one.
• Data‑Driven Evaluator & Pragmatist: You treat data as a first‑class component of the system, design robust evaluation datasets, and rigorously benchmark alternatives to select solutions based on measurable trade‑offs.
• System‑Level Collaborator: You think beyond the model and understand how ML components interact with robotics, control logic, and distributed AV systems.

Responsibilities

• Evaluate and benchmark state‑of‑the‑art ML models and algorithms for perception, tracking, and multimodal awareness.
• Design and maintain reproducible evaluation pipelines measuring model performance, latency, memory footprint, and robustness.
• Integrate ML models into production systems in collaboration with Robotics and Platform teams.
• Optimize inference pipelines for real‑time performance on constrained hardware (CPU/GPU/edge devices, Q‑SYS Cores).
• Improve model efficiency using quantization, pruning, distillation, and runtime optimization techniques.
• Write production‑grade Python (and C++ where appropriate) following clean architecture and modular design principles.
• Contribute to CI/CD pipelines, automated testing, regression validation, and performance monitoring for ML components.
• Ensure reproducibility, versioning, and traceability of models, datasets, and experiments.
• Collaborate to industrialize promising prototypes into scalable production systems.
• Work with Product and System Architects to align ML solutions with hardware and product roadmap constraints.

Qualifications

• MSc or PhD in Computer Science, Engineering, Robotics, or related technical field.
• 5+ years of hands‑on experience in machine learning engineering or applied ML roles.
• Proven experience integrating ML models into production systems.
• Strong proficiency in Python and modern ML frameworks (PyTorch, TensorFlow, ONNX).
• Solid software engineering fundamentals, including modular design, code reviews, testing strategies, and CI/CD.
• Experience optimizing models for real‑time or resource‑constrained environments.
• Understanding of system‑level trade‑offs in latency‑sensitive or distributed architectures.
• Ability to work independently and drive technical decisions within architectural guidelines.
• Strong communication skills and experience collaborating in cross‑functional engineering teams.
• Preferred experience with one or more of the following:
• Experience with computer vision, tracking, or multimodal perception systems.
• Experience with C++ in performance‑critical environments.
• Familiarity with AV systems, media pipelines, or robotics‑oriented architectures.
• Exposure to ROS, TensorRT, or MLOps tools (MLflow, Weights & Biases, Docker).