Role: AI Platform Engineer
Type: 12-month contract
Location: Toronto, ON (Hybrid)
Key Responsibilities
Platform Enablement & Governance
- Lead AI platform enablement workstreams, enable upcoming AI features and products for the organization, onboarding users/teams and use cases onto enterprise AI platforms (e.g. OpenAI, Gemini Anthropic, Amazon Bedrock, LLM gateways, and agentic frameworks).
- Coordinate with cross-functional stakeholders - Information Security, Risk, Compliance, Legal, and business teams - to review, negotiate, and agree on platform controls and guardrails.
- Translate agreed security, risk, and compliance requirements into technical controls, implemented via platform configuration changes or custom code (e.g., IAM policies, guardrails, content filters, logging/monitoring, rate limits, data-access controls). Work with cross engineering teams to enable these controls.
- Review and document controls, obtain signoffs, and maintain evidence for audit and compliance reviews.
- Ensure adherence to enterprise governance, DevSecOps protocols, and responsible AI standards across the platform.
- Manage token budgets, usage/credit limits, quotas, and rate limits across providers and teams; define allocation models that balance user productivity with cost discipline.
- Proactively monitor AI platform costs and usage; build dashboards, anomaly detection, and automated alerting to notify users and teams of unusual spend, usage spikes, or quota breaches before they become budget issues.
User Support & Enablement
- Provide day-to-day user support on credit/usage limits, quota management, and cost allocation for AI platform consumption.
- Advise users on AI usage guidance, approved patterns, and platform best practices.
- Support and troubleshoot technical questions related to skills, agents, MCP (Model Context Protocol) servers/integrations, prompt-based applications, and API usage.
- Create and maintain runbooks, FAQs, onboarding guides, and self-service documentation to scale support.
- Monitor operational metrics, usage, and incident data to drive continuous improvement, reliability, and platform adoption.