About Quin
Quin is redefining smart safety with real-time event detection and data-driven response systems that save lives. We embed our proprietary technology into helmets and wearables through B2B partnerships with global brands. With a strong foothold in motorcycling and a strategic focus on cycling and industrial safety, our technology is already trusted by top names across the U.S. and Europe.
We are a fast-moving, purpose-driven team building industry-leading technology that brings intelligence into safety gear. Every product we ship has a tangible impact, and every engineer at Quin has the opportunity to shape technology that protects lives.
About the Role
We are looking for a hands-on Machine Learning Engineer with 3–5 years of experience to join our growing R&D team. You will own the end-to-end development of ML features that analyse real-time motion and sensor data to extract meaningful patterns, detect key events, and drive actionable insights.
You will take full ownership of the lifecycle from data collection and preprocessing to model design, deployment, and continuous improvement. Your work will directly impact core product features that are used in real-world applications.
You should have strong data intuition, experience with time-series or sensor-based data, and a desire to build robust, scalable, and reliable ML systems. Collaboration with hardware, firmware, and domain experts will be key to delivering solutions that perform in real-world environments.
What You’ll Do
At Quin, you will not just build ML models. You will own features that interpret motion, respond to real-world conditions, and make our products smarter and safer.
Define and Own Features End to End
Take product ideas from concept to reality. Decide how sensor data should be collected, interpreted, and transformed into actionable intelligence. Set technical direction, performance targets, and the roadmap for your features.
Lead Data Collection and Exploration
Design and run lab experiments, field tests, and real-world data collection campaigns. Capture diverse motion patterns, edge cases, and extreme scenarios. Curate, clean, and structure datasets to create high-quality inputs for your models.
Engineer Features That Matter
Transform raw sensor data, including IMU, accelerometers, force sensors, and GPS, into meaningful features that capture dynamics, trajectories, angular velocity, and other motion signatures. Ensure your insights directly improve model accuracy, reliability, and responsiveness.
Design and Build ML Models for Real-World Motion
Develop models for pattern recognition, anomaly detection, and sensor fusion using both classical and deep learning approaches. Ensure predictions are robust, explainable, and reliable in real-world scenarios.
Deploy and Optimise for Embedded Systems
Take models from development to real-time, low-power deployment. Optimise for latency, memory, and energy consumption while collaborating with firmware and hardware teams to ensure smooth integration.
Test, Validate, and Iterate Relentlessly
Design lab and field tests that push systems to the limits. Measure model performance, reduce false positives, calibrate thresholds, and iterate to deliver robust features under all conditions.
Collaborate Across Teams
Work closely with hardware, firmware, product, and UX teams to ensure ML features are fully integrated, scalable, and user-ready. Communicate technical findings and actionable insights clearly to inspire the team.
See Your Work Come Alive
Every experiment, model, and optimisation you contribute will ship in real-world products. You will see the impact of your work on riders’ experiences and know that your models help keep people safe.
Basic Qualifications
- B.Tech or M.S. in Computer Science, Data Science, Electrical, Electronics, Robotics, or a related field
- 3–5 years of experience in machine learning, applied AI, or data science projects involving time-series or sensor data
- Strong proficiency in Python and ML/data libraries such as scikit-learn, Pandas, NumPy, TensorFlow, or PyTorch
- Experience working with real-time or time-series sensor data (IMU, accelerometers, force sensors, GPS)
- Experience in feature engineering and signal processing, including filtering, FFTs, transformations, and motion-based features
- Hands-on experience with data collection, cleaning, labelling, and preprocessing pipelines
Good to Have:
- Knowledge of motion kinematics, motion analysis, or physical sensor-based motion data
- Experience in human activity recognition, gait analysis, or motion pattern recognition
- Experience with sensor fusion, multimodal data integration, and physics-informed modelling
- Experience deploying ML models to embedded or edge devices, including C/C++ or low-power optimisation
- Experience with classical and deep learning models for sequential/time-series data (RNNs, CNNs, Transformers)
- Familiarity with MATLAB, SciPy, OpenSim, or other motion/physics simulation tools
- Proven ability to own ML features end-to-end, from definition and data collection to deployment and continuous improvement
