Job Title

About the Company:We build an AI-powered tool that automates the quoting process for window and door industry participants. Our platform ingests drawings, emails, and other quotes to generate accurate estimates in minutes, not days. By streamlining manual workflows, we help our customers and other stakeholders save time, reduce labor costs, and scale their businesses more efficiently.About the Role: We’re seeking a Data Engineer with strong computer vision knowledge and a solid foundation in backend development to join our AI/ML team. In this role, you’ll support the end-to-end lifecycle of computer vision models—from data ingestion and preprocessing to deployment and serving—while also helping to build and maintain the backend infrastructure that powers our AI features.Key Responsibilities:Develop and manage data pipelines for handling images used in training and inference.Understand and extend Python-based computer vision scripts for preprocessing, model evaluation and model development.Deploy deep learning models using PyTorch and serve them in production using FastAPI, Docker, and related tools.Design and maintain backend services that integrate ML models into products, with a focus on performance, scalability, and reliability.Work with ML engineers and product developers to ensure seamless integration of models into customer-facing features.Qualifications:Strong proficiency in Python, especially for data processing, computer vision tasks, and backend development.Hands-on experience with PyTorch and model training workflows.Proven experience building and maintaining FastAPI or similar web backends for serving machine learning models.Familiarity with Docker, REST APIs, and basic DevOps practices for deploying and monitoring services.Solid understanding of data structures and engineering best practices for handling large-scale visual dataBonus Skills:Experience with image annotation platforms (e.g., CVAT, Roboflow).Proficiency with MLflow for experiment tracking and model versioning.Experience training and deploying models using platforms like Amazon Bedrock, Databricks, or similar cloud-based ML services.Knowledge of MLOps best practices and model monitoring in production environments.Familiarity with cloud infrastructure (AWS, GCP) for scaling services and storage.Other:Full-time; 4-day in office (Markham, Ontario)Language Preference: English, Mandarin