AI Engineer

About the Company:MagickTech is a forward-thinking technology company specializing in Software Product Development, Mobile Applications, SaaS Solutions, eCommerce Platforms, Enterprise Systems, and System Application Development. We leverage cutting-edge technologies, including Augmented Reality (AR), Virtual Reality (VR), Artificial Intelligence (AI), Machine Learning (ML), and Data Science, to develop innovative, high-performance solutions that drive business transformation.As a part of Magick Group, MagickTech operates alongside industry leaders like MagickWoods, one of North America’s premier manufacturers and marketers of high-quality kitchen and bath furniture. Headquartered in Toronto, Ontario, Canada, MagickWoods has a strong global presence, with manufacturing and assembly facilities in India and the U.S., supported by a robust supply chain across Mexico, China, Indonesia, Vietnam, and Europe.Position Overview:We are seeking an innovative AI Engineer with expertise in AI/ML & Computer Vision to revolutionize automated space planning and design. You will build intelligent systems that automatically generate optimal interior layouts, particularly kitchen designs, by combining computer vision, 3D modelling, spatial reasoning, and rule-based design principles. The ideal candidate will bridge the gap between AI/ML technologies and interior design expertise to create automated design solutions.Key Responsibilities:Design Rule Engine Development: Build comprehensive rules engines for interior design constraints (work triangle optimization, ergonomic guidelines, building codes, aesthetic principles)Automated Kitchen Design: Develop systems that analyze room dimensions and automatically place cabinets, appliances, and accessories following design best practices3D Space Planning: Create algorithms for optimal placement of hobs, sinks, refrigerators, and storage solutions based on kitchen workflow principlesSpatial Constraint Solving: Implement complex constraint optimization for furniture placement, clearance requirements, and accessibility standardsComputer Vision for Space Analysis: Process room scans, floor plans, and measurements to understand spatial constraints and opportunities3D Asset Population: Automatically populate 3D spaces with appropriate furniture, fixtures, and appliances from product catalogsDesign Validation: Ensure generated designs meet safety codes, ergonomic standards, and functional requirementsMachine Learning for Design Preferences: Build models to understand and predict user design preferences and style choicesMulti-constraint Optimization: Balancing aesthetics, functionality, cost, and space utilizationReal-time 3D Visualization: Interactive systems for design preview and modificationRequired Technical Skills:Computer Vision & 3D ProcessingOpenCVOpen3DDetectron2/SAM/similar technologiesMachine Learning & Deep LearningPyTorchHugging Face TransformersLightGBM/XGBoostMLflowData ManagementRequired Qualifications:Bachelor's or Master's degree in Computer Science, Machine Learning, Computer Vision, or related field3+ years of experience in computer vision and machine learningStrong programming skills in PythonExperience with constraint programming and optimization problemsProven track record of building rule-based systems for complex decision-makingStrong understanding of 3D geometry, spatial relationships, and collision detectionExperience with version control (Git) and collaborative developmentPreferred Qualifications:PhD in Computer Vision, Machine Learning, or related fieldInterior Design Background: Formal training or significant experience in interior design, architecture, or space planningExperience with CAD software (AutoCAD, SketchUp, Revit) or 3D design toolsKnowledge of kitchen and furniture industry standards and suppliersExperience with cloud platforms (AWS, GCP, Azure) for ML workloadsKnowledge of containerization (Docker) and orchestration (Kubernetes)Experience with real-time 3D rendering and visualizationFamiliarity with furniture/product catalog management systemsExperience with generative design algorithms