21 Nov
|
Raas Infotek
|
Toronto
21 Nov
Raas Infotek
Toronto
Location: onsite, TorontoWe have an urgent opening for the position below.Key Skill:D365 CE, Dataverse, Power Platform, Python, GitHub Copilot/CLI, Understanding of LLMsExperience: 8+ years in software engineering, including hands‐on work building AI agents or LLM‐powered solutionsRole SummaryWe are looking for an AI Agent Developer to design, build and run reliable AI agents that automate business processes on our Dynamics 365 and Dataverse platform. The person in this role will turn business requirements into agents that work dependably in production, connected to APIs, tools and enterprise data.Our team builds with AI‐assisted engineering every day, using GitHub Copilot and the GitHub Copilot CLI. We value practical engineering and problem‐solving over expertise in any single agent framework. The ideal candidate understands how agents behave, where they fail and how to make them dependable.Key ResponsibilitiesAgent Design and DevelopmentTurn business requirements into transparent, predictable and reliable agent behavior.Build AI agents and agentic workflows using GitHub Copilot (IDE, Chat and Agent mode) and the GitHub Copilot CLI.Design multi‐step workflows, manage context and handle failures through retries, fallbacks and timeouts.Write and refine prompts, instructions and tool descriptions, with a clear understanding of LLM limitationsIntegration with Tools and DataConnect agents to APIs, tools and data sources, including the Dataverse Web API, Dynamics 365, Power Automate and Azure services.Build retrieval (RAG) pipelines when agents need external knowledge.Testing, Quality and SafetyTest,
debug and evaluate agent behavior for accuracy, task completion and regressions.Add human approval steps for sensitive actions and protect agents against prompt injection.Review AI‐generated code for correctness, security and performance before it is merged.Deployment and OperationsDeploy agents through CI/CD pipelines (Azure DevOps or GitHub Actions) across Dev, Test and Prod.Monitor agents in production and improve them based on real usage, errors and feedback.Documentation and CollaborationWrite clear design documents, runbooks and agent behavior specifications.Work with business stakeholders, product owners and CRM teams on requirements, UAT and rollout.Required Skills and QualificationsArea: What We Look ForLLM fundamentals: Solid understanding of LLMs, prompting and their limitationsAI-assisted development : Hands‐on, daily use of GitHub Copilot (IDE, Chat, Agent mode) and the GitHub Copilot CLITool and API integration: Experience connecting agents to APIs, tools and data sourcesWorkflow design: Designing multi‐step workflows, managing context and handling failuresTesting and evaluation: Strong problem‐solving, testing, debugging and agent‐evaluation skillsDynamics 365 / Dataverse: Hands‐on experience with Dynamics 365 CE / CRM, Dataverse and Power PlatformDeployment: Experience deploying, monitoring and improving applicationsCoding: Good coding skills in Python, TypeScript or a similar language, enough to review, fix and improve AI‐generated codeCommunication: Clear documentation and collaboration skills
📌 Ai Agent Developer (Toronto)
🏢 Raas Infotek
📍 Toronto