Data Science Architect (Computer Vision)
TOMRA Food is a multinational organisation and a leading provider of sensor‑based sorting, peeling and integrated post‑harvest solutions for the food industry.
With over 50 years of experience, TOMRA transforms global food production to maximise safety and minimise loss by ensuring Every Resource Counts.
We value innovation, passion, and responsibility.
Bring your expertise to a team that encourages risk‑taking, breakthrough thinking and a commitment to both customer success and employee well‑being.
Job Description
Lead the design and evolution of data science systems across machine and cloud platforms.
Shape the architecture that powers deep‑learning inference on TOMRA sorters and drive innovation in our cloud‑based MLOps platform, developed to train and deploy models at scale.
This role blends technical leadership, system design, and strategic foresight to deliver scalable, high‑performance solutions that create real customer value.
Responsibilities
Design vision models for real‑world food‑sorting applications.
Architect real‑time deep‑learning inference on TOMRA sorters, including GPU specifications and object‑detection, segmentation, and related tasks.
Develop and evaluate DL vision models (CNN‑based such as Res Net, Mobile Net, Efficient Net and transformer‑based architectures), selecting suitable backbones, necks, and heads.
Maintain and improve the in‑house cloud‑based MLOps platform (data‑science pipelines).
Propose new features for the MLOps platform and advise on architecture.
Stay current with Azure ML ecosystem advancements and implement efficiency improvements.
Optimize the model lifecycle, enhancing performance and creating customer value.
Produce and communicate architectural plans to the data‑science team.
Technical Leadership
Provide technical leadership and guidance to data scientists.
Track industry trends and emerging technologies to keep systems cutting‑edge.
Translate customer needs into technical product requirements.
Champion new DL model trends and library updates within the team.
Qualifications
Master's or higher degree in Artificial Intelligence, Mathematics, Statistics or an equivalent field.
Experience in a Data Science Architecture role.
Computer vision background, including multi‑modal vision (RGB, hyperspectral, NIR imaging).
Experience with DL inference on edge devices and GPU benchmarking.
Proficiency with Python, Py Torch, Tensor RT, Tensor Flow, ONNX Runtime and embedded GPU deployment.
End‑to‑end product experience: from data capture to production‑ready model deployment.
Hands‑on with Azure ML, Databricks, AWS or similar, and tools like MLFlow.
Experience with data visualisation and labeling tools (e.g., Voxel51, Label Studio).
Database knowledge (SQL, Lake FS, Azure Blob).
CUDA experience is a bonus.
Track record of delivering full development cycles for large‑scale production data‑science products.
In-depth knowledge of data‑science principles and DL backbones.
Excellent communication, interpersonal skills and ability to influence teams.
Strong problem‑solving and troubleshooting skills.
Benefits
Health insurance.
Pension scheme:
2% employee contribution, 4% employer contribution
4% employee contribution, 6% employer contribution
6% employee contribution, 8% employer contribution
25 days annual leave + additional long‑service days.
International and collaborative working environment.
Space and support to experiment with new technologies.
Social events and inclusive company culture.
Professional training and development opportunities.
Opportunity to shape a mission to reduce waste and enhance resource efficiency.
How to Apply
If you are excited by this opportunity, we would love to hear from you.
All applications will be managed with the strictest confidence.
Equal Opportunity Employer
TOMRA is proud to be an Equal Opportunity Employer.
We provide equal employment opportunities to all employees and applicants regardless of race, colour, religion, gender, gender identity, age, national origin, disability, parental or pregnancy status, marriage and civil partnership, sexual orientation, veteran status, or any other characteristics.
Reasonable accommodations will be made and provided as requested by candidates participating in any selection process.
#J-18808-Ljbffr