Computer Vision Expert

Dremian; Netherlands; A distancia; EUR 70.000 - 120.000; Jornada completa

Descripción de la vacante

Dremian is seeking a Computer Vision Expert to build the perception system for autonomous interceptor drones. You will focus on object detection, develop dataset pipelines, and evaluate multi-sensor setups including stereo vision and radar integration.

Deployment on edge hardware is a central goal. The role requires hands-on experience with neural networks (YOLO/DETR), Python, and PyTorch, with emphasis on real-time inference on embedded devices.

Formación

• Experience training and deploying object detection models.
• Strong background in neural network architectures (YOLO, DETR, etc.).
• Experience with dataset management, labeling, and augmentation.
• Proficiency in Python and PyTorch.
• Understanding of real-time inference on edge devices.

Responsabilidades

• Curate, collect, and label training datasets; build a catalog of videos and images.
• Train and optimize neural networks for real-time detection on edge hardware.
• Evaluate stereo vision and multi-camera configurations for range estimation.
• Assess radar integration for longer-range multi-sensor fusion.
• Deploy detection models on embedded devices.
• Validate detection systems in real-world flight conditions.

Conocimientos

Object detection models

Neural networks (YOLO, DETR)

Dataset management & labeling

Python

PyTorch

Edge-device inference

Educación

PhD candidate in CV

Herramientas

TensorRT

ONNX

Descripción del empleo

We're seeking a Computer Vision Expert to build the perception system for our autonomous interceptor drones. You'll focus on object detection (detection-first approach), build and manage dataset pipelines, and evaluate multi-sensor setups including stereo vision, multiple cameras, and radar integration. This role combines neural network expertise with practical deployment on edge hardware.

Key Responsibilities

• Curate, collect, and label training datasets; build a catalog of videos and images
• Train and optimize neural networks for real-time detection on edge hardware
• Evaluate stereo vision and multi-camera configurations for range estimation
• Assess radar integration for longer-range multi-sensor fusion
• Deploy detection models on embedded devices
• Validate detection systems in real-world flight conditions

Requirements & Qualifications

• Experience training and deploying object detection models
• Strong background in neural network architectures (YOLO, DETR, etc.)
• Experience with dataset management, labeling, and augmentation
• Proficiency in Python and PyTorch
• Understanding of real-time inference on edge devices

Preferred:

• PhD (or candidate) in computer vision or related field
• Experience with stereo vision or multi-camera systems
• Knowledge of radar-camera fusion
• Experience with model optimization (TensorRT, ONNX, quantization)
• Background in drone or robotics perception

Technologies & Skills

• Tiny, agile startup: your ideas and work have direct impact
• Lots of freedom to explore and propose solutions
• Challenging, meaningful tasks at the frontier of AI and robotics
• If you solve the task, your solution will be deployed on real drones
• See your work make a real-world difference

Flexible Applications Welcome

We welcome candidates with diverse skill combinations. If you have expertise in multiple areas (e.g., computer vision + machine learning, computer vision + software development, or any other combination), we're particularly interested. We value multi-talented individuals who can contribute across different domains.

Non-paid position ideal for gaining experience and contributing to a meaningful cause

Paid Positions

Competitive per-project compensation for part-time/full-time roles

Access to research resources and publication opportunities

All positions subject to NDA and non-compete agreements