Perception Engineer (Computer Vision & Edge AI) | Zürich - #55756

About The Hiring Process

When you apply, our AI-powered matching platform considers you not only for this opportunity, but also for other relevant roles across our network, so one application can unlock multiple career opportunities.

We only work with real companies hiring for live positions. Every opportunity on our platform is an active search. If there's a strong match, we'll guide you through the process. If not, we won't ignore you.

About The Company

Our client is an ambitious AI robotics startup building the intelligent software layer powering the next generation of autonomous physical security systems.

Their mission is to enable robots and intelligent sensors to perform dangerous and critical security tasks by combining cutting-edge AI, robotics, and computer vision. Backed by experienced founders and leading investors, they're building production-ready AI systems that operate in the real world—not research demos.

This is an opportunity to join one of Europe's most exciting robotics startups at an early stage and help shape products deployed in real-world environments.

• Location: Zürich, Switzerland (On-site)
• Employment Type: Full-time
• Level: Mid-Senior Level
• Visa Sponsorship & Relocation Support: Available across Europe
  
  
  

About The Role

We're looking for a Perception Engineer to build production-grade computer vision systems powering intelligent robotic platforms.

This is a highly hands-on engineering role focused on designing, deploying, and optimizing real-time perception systems—not publishing research papers. You'll work across object detection, multi-object tracking, cross-camera re-identification, video pipelines, edge AI optimization, and modern vision-language models.

You'll collaborate closely with robotics, AI, and software engineers to build systems that perform reliably in real-world deployments.

If you enjoy solving difficult perception challenges and seeing your models operate on real robots in the field, we'd love to hear from you.

What You'll Do

Computer Vision;

• Build production-ready object detection, tracking, and cross-camera re-identification systems
• Develop high-performance multi-camera perception pipelines
• Integrate modern Vision-Language Models (VLMs) for scene understanding and open-vocabulary detection
  
  
  

Edge AI;

• Deploy and optimize deep learning models for NVIDIA Jetson and other edge hardware
• Improve inference performance using TensorRT, quantization, pruning, and latency optimization
• Ensure perception systems meet real-time performance requirements
  
  
  

Video Infrastructure;

• Build scalable RTSP and WebRTC video pipelines
• Handle multi-stream processing, synchronization, latency, and reliability
• Optimize end-to-end video ingestion and processing
  
  
  

Machine Learning;

• Evaluate, fine-tune, and deploy state-of-the-art open-source and commercial models
• Train custom models when product requirements demand it
• Build robust data pipelines for image and video datasets
• Design reproducible evaluation workflows and performance monitoring
  
  
  

Cross-Functional Collaboration;

• Work closely with Robotics, AI, and Software Engineering teams
• Integrate perception outputs into autonomous robotic systems
• Help shape the long-term perception architecture of the platform
  
  
  

What We're Looking For

Must Have;

• 3+ years of experience building production computer vision or machine learning systems
• Strong Python and PyTorch (or JAX) experience
• Experience with object detection and multi-object tracking
• Experience deploying ML models on edge devices
• Hands-on TensorRT optimization and model acceleration
• Experience with quantization, latency optimization, or pruning
• Knowledge of RTSP and/or WebRTC video streaming
• Experience building scalable image/video data pipelines
• Strong software engineering fundamentals
• Ability to take systems from prototype to production
  
  
  

Nice to Have;

• Cross-camera ReID experience
• Vision-Language Models (VLMs)
• NVIDIA DeepStream
• MLOps and ML deployment pipelines
• CI/CD for Machine Learning
• Model monitoring
• Robotics or autonomous systems
• Drone or surveillance systems
• RAG and LLM integrations
• ROS2
• Publications at CVPR, ICCV, NeurIPS, ICLR, etc.
  
  
  

What We Offer

• Competitive salary
• Visa and relocation support across Europe
• Work alongside world-class AI, Robotics, and Physics founders
• Access to large-scale compute resources and real-world robotic data
• High ownership from day one
• Small, ambitious engineering team with minimal bureaucracy
• Opportunity to build production AI systems deployed in real-world security environments
• Modern office in Zürich
• International, collaborative, and engineering-driven culture
  
  
  

If you're passionate about deploying state-of-the-art computer vision systems into real-world robotics applications and want to help shape the future of autonomous security, we'd love to hear from you...