Computer Vision / Machine Learning Engineer

ManoMotion; Stockholms kommun; On-site; SEK 900,000 - 1,400,000; Full time

Job summary

ManoMotion is seeking a Computer Vision Engineer to craft robust, production-ready vision systems. You will blend ML with image processing, tracking, and geometry to boost accuracy, latency, and reliability while building end-to-end customer solutions.

You will collaborate with AI engineers on model architecture, training, and datasets, and you will own deployment, validation, and performance tuning across edge devices and constrained hardware.

Qualifications

• Master’s degree or equivalent professional experience in Computer Science, CV, EE, Robotics or ML.
• At least 3 years of professional experience developing computer-vision or AI-based products.
• Strong practical knowledge of image processing, object detection, segmentation, tracking, motion analysis, geometry, and camera systems.
• Good understanding of deep learning, neural-network training, model evaluation, and data management.
• Proven ability to combine machine-learning models with traditional CV algorithms and system-level logic.
• Strong programming skills in Python and experience with OpenCV, NumPy, PyTorch, TensorFlow, or similar frameworks.
• Good knowledge of C++ and high-performance software development.
• Experience deploying and optimizing models using ONNX, TensorRT, CUDA, OpenVINO, or similar tools.
• Experience developing real-time vision systems for edge devices or constrained hardware.
• Strong debugging, problem-solving, and software-engineering skills.
• Ability to work independently and take ownership of complex technical solutions.
• Excellent English communication skills and confidence interacting directly with customers.

Responsibilities

• Design complete computer-vision solutions combining AI/ML, image processing, tracking, geometry, and rule-based algorithms.
• Develop algorithms for image analysis, object detection, segmentation, motion analysis, pose estimation, and tracking.
• Build pre-processing and post-processing layers around AI models to improve accuracy, stability, and robustness.
• Analyse failure cases and determine whether they should be addressed through model improvements, additional data, tracking, filtering, image processing, or system-level logic.
• Work closely with AI Engineers responsible for model architecture, training, testing, and datasets.
• Integrate trained models into reliable, real-time computer-vision systems.
• Optimize vision pipelines for latency, frame rate, memory usage, and edge hardware.
• Take end-to-end technical ownership of customer solutions and integrations.
• Participate in customer meetings, technical workshops, demonstrations, and on-site validation.
• Build maintainable, testable, and production-quality software.
• Use modern AI-assisted development tools to accelerate development, testing, debugging, and documentation.
• Stay updated on advances in CV, ML, and AI-assisted software development.

Skills

Python

OpenCV

PyTorch

C++

CUDA

Object tracking

Image processing

Pose estimation

Real-time vision

Geometric reasoning

Education

Master's degree in Computer Science / CV / related field

Tools

Docker

CI/CD

MLOps

NVIDIA Jetson

ONNX

Job description

About the Role

ManoMotion is looking for a highly skilledComputer Vision Engineerto develop robust, high-performance vision solutions for real-world applications.

Our AI Engineers manage the day-to-day work related to neural-network architecture, training, testing, and data management. This role will work closely with them while focusing more deeply on the complete computer-vision system.

You will combine machine learning with advanced image analysis, detection, tracking, geometry, filtering, and algorithmic methods to improve the accuracy, stability, latency, and reliability of our products.

We are looking for someone who understands that a production-ready vision solution often requires more than a neural network. You should be able to identify model limitations and build additional processing, tracking, filtering, calibration, and decision logic around the AI models to achieve a reliable final product.

You will also take technical ownership of customer solutions, from understanding requirements and designing the system to implementation, deployment, and validation.

Responsibilities

• Design complete computer-vision solutions combining AI/ML, image processing, tracking, geometry, and rule-based algorithms.

• Develop algorithms for image analysis, object detection, segmentation, motion analysis, pose estimation, and tracking.

• Build pre-processing and post-processing layers around AI models to improve accuracy, stability, and robustness.

• Analyse failure cases and determine whether they should be addressed through model improvements, additional data, tracking, filtering, image processing, or system-level logic.

• Work closely with AI Engineers responsible for model architecture, training, testing, and datasets.

• Integrate trained models into reliable, real-time computer-vision systems.

• Optimize vision pipelines for latency, frame rate, memory usage, and edge hardware.

• Take end-to-end technical ownership of customer solutions and integrations.

• Participate in customer meetings, technical workshops, demonstrations, and on-site validation.

• Build maintainable, testable, and production-quality software.

• Use modern AI-assisted development tools such as Cursor, Claude, ChatGPT, Copilot, and other relevant tools to accelerate development, testing, debugging, and documentation.

• Stay updated on advances in computer vision, machine learning, and AI-assisted software development.

Key Qualifications

• Master’s degree, PhD, or equivalent professional experience in Computer Science, Computer Vision, Electrical Engineering, Robotics, Machine Learning, or a related field.

• At least 3 years of professional experience developing computer-vision or AI-based products.

• Strong practical knowledge of image processing, object detection, segmentation, tracking, motion analysis, geometry, and camera systems.

• Good understanding of deep learning, neural-network training, model evaluation, and data management.

• Proven ability to combine machine-learning models with traditional computer-vision algorithms and system-level logic.

• Strong programming skills in Python and experience with OpenCV, NumPy, PyTorch, TensorFlow, or similar frameworks.

• Good knowledge of C++ and high-performance software development.

• Experience deploying and optimizing models using technologies such as ONNX, TensorRT, CUDA, OpenVINO, or similar tools.

• Experience developing real-time vision systems for edge devices or computationally constrained hardware.

• Strong debugging, problem-solving, and software-engineering skills.

• Ability to work independently and take ownership of complex technical solutions.

• Excellent English communication skills and confidence interacting directly with customers.

Preferred Qualifications

• Experience with object tracking, temporal filtering, sensor fusion, or multi-camera systems.

• Experience with camera calibration, 3D geometry, depth estimation, or pose estimation.

• Experience with NVIDIA Jetson or other embedded GPU platforms.

• Experience in industrial, robotics, automotive, safety, or human-machine interaction applications.

• Experience working with challenging real-world conditions such as occlusion, poor lighting, motion blur, and changing environments.

• Experience with Docker, CI/CD, cloud platforms, MLOps, or model monitoring.

The Profile We Are Looking For

You are a strong problem-solver who enjoys turning advanced computer-vision technology into dependable products.

You understand AI models and training, but your main strength is designing the complete system around the model. When the neural network alone is not enough, you know how to add tracking, filtering, image processing, geometric reasoning, calibration, and decision logic to make the final product work reliably.

You are comfortable working across algorithms, software, hardware constraints, and customer requirements, and you take ownership of the overall result.