AI Engineer – Generative AI & Agentic Systems
- Market p25–p75
- CAD 135k–CAD 215k
- Opened · 30d
- +153
- Closed · 30d
- −148
- Signal
- ● Steady+5 net
- Opened · 7d
- +2779% of open
- All ml / ai
- 6,573open
Build the next generation of AI-enabled insurance workflows, combining intelligent automation, enterprise engineering and human decision-making.
The Role
At Capco, we're helping financial services organizations modernize complex technology environments and deliver meaningful business transformation.
We're looking for an AI Engineer – Generative AI & Agentic Systems to support an exciting engagement with Sun Life, reimagining its Individual Insurance New Business workflow and establishing a modern, scalable foundation for future transformation.
You'll design, build, test and deploy AI-enabled and agentic capabilities that integrate with workflow orchestration, case management, decision services and enterprise systems. Working alongside business and technical specialists, you'll develop secure, production-ready solutions that combine AI with deterministic rules, human oversight and robust governance.
This is a hands-on engineering opportunity to turn complex insurance workflows into practical, reliable technology solutions within a regulated environment.
What You’ll Do
Build and deploy AI-enabled solutions: Develop AI and agentic services using Python, approved cloud platforms and enterprise engineering practices, incorporating retrieval, structured outputs, tool calling, orchestration and state management.
Engineer enterprise integrations: Design APIs, asynchronous services and integrations connecting AI capabilities with workflow engines, case-management platforms, rules services, document processing and existing systems of record.
Develop intelligent document and evidence processing: Implement document classification, extraction, reconciliation and information retrieval capabilities to support insurance assessment, triage and operational workflows.
Embed responsible AI and engineering controls: Build evaluation datasets, automated regression tests, human-review workflows, authorization controls and guardrails that support secure, auditable and reliable AI operations.
Deliver production-ready capabilities: Implement CI/CD pipelines, monitoring, logging, tracing and operational documentation while collaborating with domain specialists to test normal, exception and failure scenarios.
What We’re Looking For
Software and AI engineering experience: 3+ years in professional software, data or ML engineering, including 1+ year of hands-on AI/ML engineering and practical experience delivering Generative AI, LLM or agentic solutions.
Python and enterprise integration expertise: 2+ years of Python development and 2+ years building APIs, asynchronous services or enterprise integrations, with knowledge of FastAPI or equivalent frameworks, SQL and structured/unstructured data processing.
Cloud and production engineering capabilities: 2+ years working with AWS, Azure or GCP, alongside experience with Git, Docker, CI/CD, automated testing, secure coding, secrets management and production troubleshooting.
Applied AI engineering knowledge: Experience with LLM APIs, structured outputs, tool/function calling, retrieval-augmented generation, embeddings, vector retrieval, agent orchestration, state and memory patterns, prompt/configuration versioning and AI evaluation.
Collaborative delivery and governance mindset: Analytical problem-solving, clear communication, effective documentation and the ability to work iteratively with stakeholders, alongside an understanding of privacy, security, auditability, workflow decisioning and human oversight.
Bonus Points For
Experience with Amazon Bedrock, Bedrock AgentCore, LangGraph, Google ADK, Semantic Kernel or comparable managed AI and agent orchestration frameworks.
Familiarity with Model Context Protocol (MCP), Neo4j, graph databases, agent-tool interoperability or evaluation of agent outputs and execution trajectories.
Experience with document intelligence, workflow and case-management platforms, event-driven architectures or durable messaging.
Knowledge of legacy-system integration, mainframe environments, infrastructure as code or enterprise modernization programmes.