AI Automation Engineer - CoPilot

CGI•Canton, MA

2d•Remote

About The Position

CGI is seeking an AI Automation Engineer to design, deploy, and support intelligent AI agents in Microsoft Copilot Studio that transform manual tasks and streamline business processes. In this highly collaborative role, you’ll work closely with the AI Architect and business stakeholders to identify process challenges, turn ideas into practical AI-powered solutions, and measure their adoption and business impact. This position can be performed near any CGI office (preference in MA).

Requirements

Must have at least 6 + years of experience as a full stack developer with Python or similar language
3+ years in automation, software, or solutions engineering
Hands on experience with Microsoft Copilot Studio and Power Platform (Power Automate, Power Apps, Dataverse) - Mandatory
Working knowledge of generative AI concepts: LLMs, prompt engineering, grounding/RAG, and agent limitations
Ability to translate business processes into automated workflows
Strong communication skills and comfort working directly with non technical stakeholders

Nice To Haves

Microsoft certifications (PL 200, PL 400, AI 900, AI 102)
Experience with Azure AI services (Azure OpenAI, AI Foundry, AI Search)
Familiarity with agent frameworks beyond Copilot Studio
Basic Python or JavaScript for custom actions
Power BI for usage and ROI reporting
Experience in healthcare, insurance, or other regulated industries

Responsibilities

Requirements discovery: Work with the AI Architect and business users to understand manual workflows, identify automation candidates, and define agent scope and success criteria.
Agent development: Build agents in Copilot Studio, including topics, prompts, knowledge sources, actions, and conversation flows.
Workflow automation: Connect agents to systems using Power Automate, connectors, and APIs (e.g., SharePoint, Teams, Outlook, Dataverse, Excel, ServiceNow, internal apps).
Prompt and knowledge design: Write and refine instructions, ground agents on enterprise content, and tune responses for accuracy and consistency.
Testing and quality: Test agents for accuracy, edge cases, and failure handling. Validate with business users before release.
Deployment and governance: Follow architect defined standards for security, data access, DLP, environment management (dev/test/prod), and responsible AI.
Usage tracking and reporting: Monitor agent adoption, volume, resolution rates, and time/effort saved. Build simple dashboards (Power BI or Copilot Studio analytics) and report progress to leadership.
Continuous improvement: Review feedback and usage data, then iterate on agents to improve performance and expand coverage.
Documentation and enablement: Maintain agent documentation, runbooks, and user guides. Train business users on adoption.