AI/ML LLM Engineer

Akkodis is seeking an LLM Engineer for a contract position with a client in Toronto, ON (Hybrid). Ideally looking for experience with implementing evaluation frameworks for GenAI applications, AWS cloud services, containerization (Docker, Kubernetes), RAG and serverless architectures, at least 3 years in a full-stack or backend-heavy role. preference would be someone with the required skills and experience, particularly in large organizations.

Title: LLM Engineers

Location: 22 Adelaide Street W Ste 2720 c M5H 4E3 Canada

Qualifications

• 6+ years of professional software development experience, with at least 3 years in a full-stack or backend-heavy role.
• Strong knowledge of AWS cloud services, containerization (Docker, Kubernetes), and serverless architectures.
• 2+ years in LLM application development and modern LLM-enabled coding practices. Experience in end-to-end implementations of RAG architectures, document embedding/indexing pipelines, and agentic workflows.
• Familiarity with GenAI evaluation methodologies, including prompt testing, benchmark datasets, and performance monitoring.
• Solid grasp of frontend architecture (state management, performance, testing, accessibility) and integration with APIs/backends.
• Strong Agile practice: story slicing, definition of ready/done, metrics in Jira, and pragmatic ceremony management.
• Experience in life sciences industry is desirable, particular pharmaceutical research and clinical development.

Role summary

• Independently contribute to the development, testing, and deployment of custom GenAI applications in an enterprise environment. Applications are in support of drug discovery research, but domain experience is not essential.
• full-stack applications leveraging GenAI models and enterprise integration patterns.
• Design and implement evaluation frameworks for GenAI applications, including automated prompt testing, regression testing, and human-in-the-loop review workflows.
• Build and optimize APIs, microservices, and front-end components for high performance and usability.
• Track delivery health in Jira (throughput, blockers, rework) and report concise status to stakeholders.