Edge Vision Engineer for Real-Time Robotics

TechTree; Zürich; Remote; CHF 140,000 - 180,000; Full time

Benefits offered by this job

In-person Zürich

Remote considered

Autonomy

Strong founding team

Job summary

TechTree is building the perception stack powering multi-sensor surveillance and robotics in Zurich. You will ship cross-camera ReID and high-throughput detection pipelines, integrating vision-language models for open-vocabulary understanding.

The role emphasizes reliability, real-time performance, and deployment on edge hardware. You will own ingestion, streaming (RTSP/WebRTC), and latency optimization while collaborating with robotics and software teams to ensure perception outputs support

Qualifications

• Proven track record shipping computer vision or ML systems in production (3+ years or equivalent depth).
• Strong understanding of SOTA techniques for cross-camera ReID, object detection, and multi-object tracking.
• Hands-on experience with VLMs.
• Real-time video streaming experience with RTSP/WebRTC.
• Experience deploying on edge hardware: TensorRT, quantization, latency tuning.
• Strong Python and PyTorch (or JAX) skills.
• Broad ML literacy and software engineering hygiene.
• Strong data pipeline skills: curating, cleaning, labelling, and managing large image and video datasets.

Responsibilities

• Design and ship cross-camera ReID for people and vehicles across distributed camera networks.
• Build high-throughput object detection and multi-object tracking pipelines across many video streams.
• Integrate and adapt vision-language models for open-vocabulary detection and scene understanding.
• Own the video ingestion and streaming path from camera to model, focusing on latency and resilience.
• Optimize and deploy models on edge hardware (Jetson) with TensorRT and pruning.
• Evaluate, fine-tune, and integrate existing models and train custom models when needed.
• Work closely with software and robotics teams so perception feeds autonomy and alerting.

Skills

Computer vision

Edge deployment

Python PyTorch

RTSP WebRTC

NVIDIA DeepStream

TensorRT

VLMs

ML systems in prod

Tools

TensorRT

NVIDIA DeepStream

PyTorch

JAX

Job description

TechTree is building the perception stack powering multi-sensor surveillance and robotics in Zurich. You will ship cross-camera ReID and high-throughput detection pipelines, integrating vision-language models for open-vocabulary understanding.

The role emphasizes reliability, real-time performance, and deployment on edge hardware. You will own ingestion, streaming (RTSP/WebRTC), and latency optimization while collaborating with robotics and software teams to ensure perception outputs support