Applied AI Software Engineer

Position Summary:

The Applied AI Software Engineer is a hands-on technical role responsible for designing, building, testing, deploying, and supporting secure, reliable artificial intelligence solutions within SE Health. The role translates business, clinical, and operational needs into production-ready applications using machine learning, generative AI, natural language processing, retrieval-augmented generation, agentic workflows and orchestration, automation, and related technologies.

Working with product, enterprise architecture, data, security, privacy, clinical, and software engineering partners, the Applied AI Software Engineer delivers solutions across the full development lifecycle, from discovery and prototyping through integration, release, monitoring, and continuous improvement. The role also applies modern agentic software development practices and AI coding agents under human review to increase delivery speed and consistency without compromising quality, security, or accountability. The position applies strong software engineering practices, responsible AI controls, and user-centred design to ensure AI capabilities are explainable, maintainable, cost-effective, and aligned with SE Health standards and strategic priorities.

Primary Accountabilities:

AI Solution Design & Development

• Translate business, clinical, and operational requirements into secure, maintainable AI solution designs and technical specifications.
• Develop AI-powered applications, services, agents, and APIs using Python and TypeScript/Node.js, and approved frameworks, models, and cloud services.
• Build retrieval-augmented generation solutions, including document ingestion, chunking, embeddings, vector search, grounding, prompt design, and response orchestration.
• Select, configure, and integrate machine learning and foundation models based on quality, risk, privacy, latency, scalability, and cost requirements.
• Create reusable components, implementation patterns, and technical documentation that accelerate consistent AI delivery.
• Participate in architecture, design, and code reviews and align solutions with enterprise technology standards.
• Design and build multi-stage large language model pipelines and hybrid retrieval combining vector and lexical search, with section-aware chunking, re-ranking, citation-level source attribution, checkpointing, resumability, and human-in-the-loop approval points.
• Implement structured, schema-validated model outputs and tool calling, and build document ingestion, generation, redlining, and change-summary capabilities across Word, PDF, and web sources.         
   

Software Engineering, Integration & Delivery

• Build and maintain data pipelines, model integrations, REST APIs, and application components required to operationalize AI capabilities.
• Apply version control, automated testing, peer review, continuous integration and delivery, containerization, and infrastructure automation practices.
• Integrate AI services with enterprise applications, workflows, identity controls, data platforms, and approved cloud environments.
• Develop unit, integration, regression, performance, and safety tests; troubleshoot defects and resolve production issues.
• Collaborate in agile delivery with product owners, designers, data professionals, architects, security specialists, quality engineers, and operational teams.
• Build full-stack application components in TypeScript and Node.js, including React and Next.js or equivalent, as well as Python, backed by relational databases with row-level security, versioned schema migrations, and immutable audit trails.

• Apply agentic software development methodologies and AI coding agents, such as Claude Code, OpenAI Codex, GitHub Copilot, or equivalent, including specification-driven workflows, autonomous iteration loops, repository-level agent instruction files, custom commands, skills and hooks, subagent orchestration, Model Context Protocol servers, and tiered automated validation gates, with all agent-generated work peer reviewed, tested, and traceable in source control.

   

Responsible AI, Security & Quality

• Embed privacy, security, accessibility, human oversight, transparency, and responsible AI requirements throughout the development lifecycle.
• Design evaluation datasets and automated checks for accuracy, relevance, groundedness, hallucination, bias, toxicity, robustness, and task completion.
• Protect sensitive health and personal information through approved data handling, access controls, logging, encryption, and secure development practices.
• Document model choices, data lineage, limitations, risks, controls, testing evidence, and operational procedures to support governance and auditability.
• Partner with privacy, security, legal, clinical safety, and risk stakeholders to resolve issues before release.
• Build evaluation and guardrail infrastructure for generative and agentic workflows, including golden datasets, deterministic regression suites, retrieval and recall metrics, rubric scoring, model-as-judge assessors, least-privilege tool and data access, prompt-injection defences, sandboxed execution, and auditable human approval gates before any consequential action.
   

Operations, Evaluation & Continuous Improvement

• Deploy and operate AI solutions in non-production and production environments using established DevSecOps and MLOps practices.
• Implement monitoring and alerting for model quality, drift, latency, availability, token consumption, cost, security events, and user feedback.
• Analyze telemetry and evaluation results to tune prompts, retrieval, models, workflows, and infrastructure.
• Support incident response, root-cause analysis, defect remediation, and controlled rollback or recovery.
• Stay current with emerging AI engineering practices and recommend practical improvements through prototypes and technical experiments.
• Implement cost and token governance for AI workloads, including budget envelopes, per-run cost ledgers, caching, batching, and model routing.
• Maintain run manifests, versioned prompts, and reproducibility evidence so AI runs can be traced, replayed, and rolled back.
   

Qualifications & Experience:

• Bachelor’s Degree in Computer Science, Software Engineering, Artificial Intelligence, Data Science, Engineering, or a related technical field.
• Equivalent combination of education, professional experience, and demonstrated technical capability will be considered. Relevant cloud, AI, machine learning, security, DevOps, or MLOps certifications are considered assets.     
• 7+ years of professional software development experience, including hands-on delivery of AI, machine learning, data, or automation solutions.
• Strong proficiency in Python and TypeScript/Node.js, source control, automated testing, API development, and SQL or data-processing workflows.
• Hands-on experience with Microsoft Foundry, including Foundry Models, Agent Service, SDKs, model deployment, prompt flows or agent workflows, evaluation, tracing, monitoring, content safety, and enterprise governance.
• Experience developing with Anthropic Claude models through the Claude Messages API, Anthropic SDKs, or Microsoft Foundry, including tool use, structured outputs, prompt caching, long-context workflows, and secure authentication using Microsoft Entra ID or API keys.
• Experience building AI or generative AI applications using large language model APIs, prompt engineering, embeddings, vector search, retrieval-augmented generation, agentic orchestration, Model Context Protocol, and model evaluation.
• Experience comparing and selecting models such as Claude, Azure OpenAI models, and other foundation models based on quality, safety, latency, context-window, data-residency, scalability, and cost requirements, including token budgeting, prompt and context caching, and model routing to manage spend and performance.
• Experience with one or more machine learning or AI application frameworks and libraries, such as PyTorch, TensorFlow, scikit-learn, Hugging Face, LangChain, Semantic Kernel, or equivalent tools.
• Experience deploying cloud-based applications or AI services using containers, continuous integration and delivery pipelines, monitoring, and secure configuration practices.
• Working knowledge of privacy, cybersecurity, responsible AI, data governance, and risk controls; healthcare or another regulated-sector context is an asset.
• Demonstrated ability to communicate technical concepts clearly, work collaboratively, and deliver maintainable solutions in an agile environment.
• Hands-on experience with agentic development methodologies and AI coding harnesses—such as Claude Code, OpenAI Codex, GitHub Copilot, Cursor, or equivalent—including repository-level agent instruction files (AGENTS.md, CLAUDE.md), custom commands, skills and hooks, subagent orchestration, Model Context Protocol servers, and specification-driven delivery workflows such as GitHub Spec Kit.

• Experience with full-stack web development (React, Next.js, or equivalent), relational databases with vector search such as PostgreSQL with pgvector, multi-stage LLM pipeline delivery with human-in-the-loop gates, generative AI evaluation infrastructure, and AI cost and token governance.

   

Competencies:

• Applied AI and software engineering
• Python and TypeScript/Node.js, APIs, data integration, and cloud development
• Microsoft Foundry, Foundry Models, Agent Service, evaluation, and observability
• Anthropic Claude models, Messages API, tool use, and long-context application development
• Generative AI, retrieval-augmented generation, agentic workflows, Model Context Protocol, prompt design, and evaluation
• DevSecOps, MLOps, automated testing, monitoring, and troubleshooting
• Privacy, security, accessibility, and responsible AI by design
• Analytical problem-solving and attention to quality
• Cross-functional collaboration and user-centred delivery
• Clear technical documentation and communication
• Agentic development methodologies and AI coding agents (Claude Code, OpenAI Codex, or equivalent) applied under human review
• Full-stack web development – React and Next.js, PostgreSQL with pgvector, AI evaluation engineering, and cost and token governance
   

Why join SE Health?

• Competitive Total Rewards
  So much more than a paycheque! Enjoy comprehensive benefits, pension, flexible pay options, car-loan support, housing solutions and exclusive staff perks.
• Flexibility & Belonging
  Thrive with hybrid work, flexible scheduling and a supportive, inclusive culture that puts people first.

• Innovative

   

  At SE, we are always looking for new, innovative ways to improve. You’ll be encouraged and supported to identify and make improvements to the way we do our work. As a social enterprise, we support research into Senior’s Health and Aging.

• Purpose & Impact
  Join a national social enterprise where your voice matters. Every role helps advance health, spark innovation and strengthen communities across Canada.

• Growth That Meets Your Ambition
  Access tuition support, training and meaningful career pathways across a growing, future-focused organization.

   

About SE Health

SE Health is a not-for-profit social enterprise advancing health with heart. With 115+ years of impact, we bring hope, happiness and exceptional care to people and communities across Canada. We lead with empathy, dignity and purpose while building a future where everyone can realize their full health and well-being potential. We’re also an inclusive, supportive workplace offering competitive compensation, strong benefits and real opportunities to grow. We’re All In Together.

Accessibility: If you require accommodations due to illness or disability, please contact Talent Acquisition at careers@sehc.com.

AI and compensation details: We use AI to take notes during our interview.  All applications and interviews are reviewed by our Talent Acquisition team. This role is a new addition. The total target compensation for this position is $87,000 – $108,000. The compensation offered is determined based on the successful candidate’s relevant experience, skills, and competencies, taking into consideration internal equity.