Robotics Software Engineer

Location: Singapore - St James Power Station Headquarters

Time Type: Full time

Job Description

About Us

Dyson is a global technology company that sets out to solve the problems others ignore. We create machines that are different, better and more useful through inventive engineering, relentless testing and a refusal to accept conventional answers.

About the Role

As a Robotics Software Engineer, you will design and develop core robotics behaviours and features for intelligent products, contributing to high-performance, reliable, and scalable robotic systems. You will work within a cross-functional team to solve real-world robotics problems, particularly in navigation, perception, and autonomous behaviour.

Key Responsibilities

• Design and develop robotics behaviours and features for embedded systems

• Implement algorithms for navigation, control, and task execution in robotic platforms (e.g., vacuum robots)

• Contribute to full software lifecycle: requirements, design, implementation, testing, and maintenance

• Participate actively in feature design discussions and code reviews

• Optimize system performance, robustness, and reliability

• Collaborate with hardware, product, and testing teams

AI/ML Responsibilities

• Implement and integrate ML models into embedded/edge robotics systems, ensuring efficient runtime performance (latency, memory, power constraints)

• Support development of perception modules:

  • Object detection and classification (e.g., obstacles, floor types, dirt detection)

  • Basic scene understanding for navigation

• Assist in data pipeline activities:

  • Data collection from robots (logs, sensor data)

  • Data annotation and preprocessing for model training

• Work on feature-level intelligence:

  • Cleaning optimization behaviours (coverage, path efficiency)

  • Basic adaptive behaviours based on sensor inputs

• Collaborate with ML engineers to:

  • Validate models in real-world robotic environments

  • Debug model failures in production scenarios

• Use GenAI tools (Copilot, Claude, etc.) to:

  • Accelerate implementation of ML pipelines and test harnesses

  • Improve code quality and documentation for AI components

About You

• 2+ years in robotics or embedded software development

• Strong programming skills in C++ (modern standards), Python

• Experience in:

  • Robotics behaviour development (navigation, path planning, control)

  • Embedded systems and real-time software

• Familiarity with software design patterns and clean architecture

• Foundational understanding of machine learning principles:

  • Supervised learning, classification, regression

  • Basic neural networks and model evaluation concepts

• Exposure to robotics perception fundamentals:

  • Computer vision (e.g., OpenCV, image processing)

  • Sensor data handling (camera, LiDAR, IMU)

• Familiarity with ML frameworks (e.g., PyTorch, TensorFlow Lite) for basic model integration

• Understanding of edge/embedded ML constraints:

  • Latency, memory, compute, and power trade-offs

• Experience supporting data pipelines:

  • Data collection, preprocessing, and labeling workflows

• Basic knowledge of robotics behaviour systems:

  • Navigation, obstacle avoidance, simple planning logic

• Awareness of performance optimization techniques:

  • Multithreading, concurrency, and efficient resource use

• Practical experience using GenAI tools (e.g., GitHub Copilot, Anthropic Claude):

  • Code generation, debugging assistance, and documentation

• Strong analytical skills with ability to debug real-world robotics + ML issues

• Strong problem-solving mindset and learning agility

Dyson is an equal opportunity employer. We know that great minds don’t think alike, and it takes all kinds of minds to make our technology so unique. We welcome applications from all backgrounds and employment decisions are made without regard to race, colour, religion, national or ethnic origin, sex, sexual orientation, gender identity or expression, age, disability, protected veteran status or other any other dimension of diversity.