What You'll Do:We are seeking an experienced and hands-on Applied AI Scientist to develop and improve the models behind one of CTC's strategic customer intelligence platforms. The platform combines machine learning, generative AI and large-scale transaction data to identify customer shopping missions and behavioural patterns across hundreds of millions of transactions. Its outputs support Personalization, marketing, merchandising, dealer and broader customer analytics use cases.
The successful candidate will focus on improving the quality of the models and their outputs. This includes developing better approaches to customer occasion discovery and transaction classification, testing new ML and GenAI methods, and demonstrating that model changes deliver measurable improvement. Working closely with ML/AI Engineers on the team, this individual will help move new approaches from experimentation into production and become a go-to technical expert on the models and how their outputs should be used.
- Develop a deep understanding of the existing engine, including occasion discovery, catalogue generation, transaction tagging, feature engineering, model inference and downstream use of model outputs.
- Develop and improve the ML and GenAI models used to identify customer shopping missions, generate occasion definitions, and classify transactions.
- Research, prototype and test new approaches across classification, clustering, and other relevant ML/AI methodologies.
- Design and improve the prompts, context strategies and model-tuning approaches used to generate, describe, and refresh customer occasions.
- Build robust evaluation frameworks for ML and GenAI outputs, including representative test sets, quality measures, human review, regression testing, and analysis of failure modes.
- Improve model accuracy, coverage, stability and interpretability, with particular focus on the quality and usefulness of the occasion catalogue and transaction assignments.
- Design experiments to compare model and prompt alternatives, quantify improvement, and determine whether changes should move into production.
- Analyze customer behavior and model outputs to identify new features, signals and modelling opportunities that strengthen Personalization and other downstream use cases.
- Translate customer and retail problems into clear modelling objectives, success measures, and technical requirements.
- Work with the ML/AI Engineers to productionize models, prompts, and evaluation processes while preserving the intended model design and performance.
- Monitor model performance over time and recommend retraining, tuning, taxonomy changes or new modelling approaches where required.
- Document key modelling decisions, assumptions, limitations and results, and explain them clearly to technical and business partners.
What You Bring:
- Graduate degree, preferably PhD, in Computer Science, Machine Learning, Artificial Intelligence, Statistics, Mathematics, Engineering or a related quantitative discipline.
- 5+ years of developing and applying ML or AI models to complex, real-world problems, including hands-on experience with GenAI and foundation LLMs.
- Advanced Python skills and strong experience with SQL, PySpark, and large-scale datasets.
- Deep understanding of supervised and unsupervised learning, including classification, clustering, feature engineering, model validation, and statistical evaluation.
- Hands-on experience with LLMs and prompt engineering, including prompt design, context construction, structured outputs, testing, and optimization.
- Experience selecting, evaluating and tuning foundation models, with a clear understanding of when to use prompt engineering, retrieval, model adaptation or fine-tuning.
- Experience evaluating GenAI systems using automated measures, human review, representative test sets, and structured error analysis.
- Experience translating ambiguous business problems into modelling approaches and explaining results, limitations and trade-offs clearly.
- Experience with Azure and Databricks, including MLflow and related model development toolkits.
- Strong technical judgement and problem-solving skills, with the ability to challenge existing methods and develop better approaches.
- Experience with Personalization, recommendation systems, customer intelligence, behavioral modelling, taxonomy development or transaction classification is an asset.
- Experience with embeddings, retrieval-augmented generation, semantic search, vector databases, or other methods for modeling unstructured data is an asset.
- Experience in retail, loyalty, marketing, ecommerce, or customer analytics is an asset.
Location: Toronto, Ontario (Hybrid: In-office 4 days per week)
We're always looking for great talent! In addition to competitive pay, we offer:
- Comprehensive benefits and retirement programs
- Performance incentives, Continuing Education Programs
- Other perks to support your well-being
- Career growth opportunities and product discounts
Broadband Salary Range: $64,000 - $106,000.
Salary decisions are also dependent on other factors such as your experience, industry benchmarks, internal equity and other role-specific requirements. For critical roles, the compensation offering will be reviewed to ensure alignment with market rate and conditions and the unique value you bring to the role.
#LI-AK1
We may use artificial intelligence tools as part of our recruitment process to assist in the initial screening of resumes. All hiring decisions, including candidate evaluation, selection, and disposition, are made by human recruiters.
