Spécialiste d'analyse de données AI-Vision par ordinateur

Waste Robotics•Montreal, QC

2d•Hybrid

About The Position

Within our AI analysts team, and in close collaboration with the AI research team and our external annotation partner, you will oversee the full lifecycle of training data: from defining classes to monitoring trained models. This is a role built on judgment as much as rigor — the core of the job is identifying, among terabytes of production images, the ones that will truly advance our models: site coverage, operating conditions, and rare classes.

Requirements

Bachelor's degree in Computer Science, Software Engineering, or a related field — or a college diploma (DEC) in Computer Science combined with equivalent experience.
Experience in data processing for machine learning or image processing for automated object classification.
Practical knowledge of Python applied to data.
Understanding of deep learning model training and evaluation processes (computer vision).
Exceptional rigor and attention to detail: in this role, data quality is non-negotiable.
Excellent communication and collaboration skills within a multidisciplinary team.
Ability to coordinate the work of an external partner: planning, deadline tracking, clear feedback loops.
French required (knowledge of English an asset).
Autonomy in organizing one's own work, and good judgment in managing priorities.
Analytical mindset and strong problem-solving skills.

Nice To Haves

Experience with annotation or data curation tools (CVAT or equivalent).
Experience with deep learning libraries (PyTorch, TensorFlow) or the YOLO ecosystem.
Experience coordinating with an external vendor or partner.
Knowledge of Agile methodologies.

Responsibilities

Optimize our clients' model classification schemas based on their sorting requirements.
Perform data pre-processing and select high-impact images and objects of interest for training (cross-site variety, operating conditions, rare classes).
Plan and coordinate the work of our external annotation partner, ensuring deadlines are met and instructions are clear.
Review and quality-control incoming annotations, and evolve the annotation guides accordingly.
Prepare training datasets and contribute to data storage management.
Launch model training runs, track results, and interpret evaluation metrics.
Contribute to improving our data curation and annotation automation tools.
Collaborate with the AI research team on continuous improvement of data and training pipelines.