Machine Learning Engineer
Pyper Vision
Location
Remote or Hybrid (Christchurch preferred)
Who we are
At Pyper Vision we are on a mission to empower humanity to work in harmony with the weather, using advanced technology to address one of the most persistent disruptions in global operations: fog and low visibility. Our goal is to enhance the resilience and operational efficiency of industries where low visibility disruption has serious impact, including aviation, maritime, and infrastructure, by pioneering world leading weather prediction and visibility forecasting technology.
Our commitment to sustainable fog management is guided by our core values:
🚀 Courageous innovation
🔍 Integrity through transparency
🌱 Lifelong stewardship
⚡ Dedicated diligence
The role
We're looking for an experienced Machine Learning Engineer with a minimum of 5 years hands-on experience building and deploying ML systems for active commercial products. If you're passionate about using advanced technology to solve real-world challenges and have proven experience with ML infrastructure, this role will be perfect for you.
The ideal candidate will have expertise in creating production-ready machine learning models, be skilled at working with complex data, and have experience building scalable ML infrastructure.
You'll be joining our close-knit product development team within our growing company of 5, where you'll help shape the technical direction and contribute across the full product lifecycle from R&D and experimentation through deploying scalable, production-ready systems. In this role, you'll play an essential part in developing a sophisticated fog prediction system, working alongside our team to focus on both model development and the infrastructure needed to deploy and maintain it in production.
Responsibilities
- Develop deep learning models for fog nowcasting, integrating diverse data sources including weather data, camera and sensor data, and forecast models
- Develop and apply advanced computer vision techniques to improve model performance
- Apply temporal analysis and time-series modelling techniques to weather data and prediction sequences
- Design, implement and maintain scalable ML infrastructure with MLOps best practices including automated training, deployment pipelines, and production monitoring
- Implement multiple learning approaches and perform rigorous model evaluation to optimise accuracy and production performance
- Collaborate closely with weather experts, domain specialists, and cross-functional peers to ensure the nowcasting model meets domain-specific requirements
- Demonstrate technical excellence across ML model development and production deployment, including robust experimentation and scalable model delivery
- Communicate effectively with team members and stakeholders, translating technical concepts for both technical and non-technical audiences
- Design and implement new ML approaches and data pipelines to support research initiatives
Requirements
- Minimum 5 years hands-on experience as a Machine Learning Engineer developing and deploying ML models for active enterprise products.
- Experience with time-series data and multi-modal learning
- Expertise in computer vision techniques and image processing
- ML Infrastructure & MLOps experience:
- Experience with ML infrastructure design and implementation (model serving, monitoring, scaling)
- Basic MLOps experience (CI/CD, experiment tracking, model versioning)
- Familiarity with containerisation (Docker, Kubernetes) and cloud-native ML deployments
- Production deployment experience: handling large datasets, deploying models in production environments, and model monitoring in live systems
- Strong experience with Python, data science tools, and deep learning frameworks (TensorFlow, PyTorch)
- Experience with Google Cloud Platform (primary) or AWS for ML workloads and data pipelines
- Strong communication and analytical problem-solving skills
Nice to have
- Experience with meteorological or atmospheric science applications
- Experience with containerisation (Docker, Kubernetes) and advanced cloud-native deployments
- Knowledge of real-time data streaming and processing systems