Responsibilities
- Apply machine learning and AI techniques including GenAI and LLM-based approaches to develop model systems and solutions, collaborating across functions to scale and integrate these into large-scale cloud-based SaaS production environments for healthcare.
- Work with product leaders, clinical informaticists, data scientists, UI/UX researchers, engineers, and healthcare professionals to design and deliver high value solutions.
- Design, build and evaluate solutions that may involve structured or unstructured data for healthcare use cases, delivering capabilities such as predictive models, summarization, recommenders, semantic search, extraction, classification, or other AI/ML applications
- Perform research and experimentation to select appropriate approaches, algorithms, and evaluation methods, and execute the R&D needed to deliver production-ready model systems.
- Perform data collection, cleaning, analysis, prompt tuning, parameter fine-tuning, model training, development, and evaluation—using existing or developing new tools or workflows as needed.
- Apply and help evolve responsible AI evaluation practices, using approaches and frameworks such as LLM-as-judge, RAGAS, and LangSmith, and design and implement custom model performance and quality metrics as needed to ensure model quality, fairness, and safety at scale.
- Contribute to a collaborative team culture, sharing knowledge and helping develop new capabilities across the organization.
Requirements
- Ph.D grad, with 1+ years related industry experience OR Masters grad with 4+ years industry experience--including at least 1 year directly related to healthcare ML/AI
- Proven industry experience through multiple major product releases in a commercial SaaS environment.
- Hands-on experience working with healthcare data (e.g. EHR, ADT, clinical notes).
- Proficiency in Python.
- Proficiency with SQL and data engineering for AI/ML applications.
- Experience working with large datasets using big data frameworks (e.g. Azure Data Lake, Apache Spark or Databricks)
- Solid understanding of transformer models and LLM-based approaches, including hands-on experience with prompt tuning and PEFT methods (e.g. LoRA, QLoRA) using frameworks such as Hugging Face Transformers.
- Experience building and evaluating models using modern ML packages such as NumPy, SciPy, Pandas, Scikit-learn, PyTorch, and LightGBM.
- Experience building and deploying models using public cloud infrastructure (Azure, AWS, or Google Cloud), including familiarity with version control, CI/CD pipelines, and scaling considerations for production ML systems at SaaS scale.
- Strong communication and collaboration skills; comfortable working on a distributed team.
Nice to Have
- Proficiency in Java or other languages helpful.
- Experience with one or more of: reinforcement learning or RLHF, NLP techniques for summarization, extraction, classification, or semantic search, retrieval-augmented generation (RAG) pipelines and/or agentic frameworks (e.g. LangChain, LlamaIndex).
Additional Information
- Travel to Office expectations For Remote Roles: If this role is remote, there will be in-office events that will require travel to and from the Mississauga and/or Salt Lake City office. These will include, but not limited to, onboarding, team events, semi-annual and annual team meetings.
- For Hybrid Roles: If this role is Hybrid, there will be an expectation to reside within commutable distance to the office/location specified in the job listing. This will include, but not limited to, weekly/bi-weekly/monthly events in the office with your specific team. This is a requirement for this role.
