AI Automation Engineer
US client via staffing partner · Irvine, CA
- Location
- Irvine, CA · Onsite
- Salary band
- Band not stated
- Type
- Full-time
- Level
- Senior
- Work authorization
- Not stated
Stack
Python · Node.js · TypeScript · Claude Code · Claude Skills · Model Context Protocol (MCP) · OAuth · SSO · Workday · ServiceNow · Microsoft 365 · SharePoint · Jira · Confluence · Slack · Excel · SQL · Power BI · Tableau · Looker
About the role
ROLE: AI Automation Engineer LOCATION: Irvine, CA (onsite) EXP: 8-12 years FULLTIME Job Description AI Automation Engineer Must Have Technical/Functional Skills ● 3+ years of software engineering experience, with solid proficiency in Python and/or Node.js/TypeScript. ● Hands-on experience building with Claude Code / Claude Skills or an equivalent agentic coding assistant (e.g., Open AI Codex, Cursor), including the skill/plugin packaging model (SKILL.md-style definitions, front-matter, tool scoping). ● Working knowledge of the Model Context Protocol (MCP) — building, deploying, and registering MCP servers; comfortable with OAuth/SSO-based auth patterns. ● Experience integrating with enterprise SaaS APIs and data sources — Workday, ServiceNow, Microsoft 365/SharePoint, Atlassian (Jira/Confluence), Slack — plus strong Excel/spreadsheet automation (formulas, pivots, reconciliation logic). ● Understanding of data classification and AI governance concepts (PII handling, RBAC, guardrails) in an enterprise setting. ● High discretion with confidential material — org-change plans, headcount, and vendor financials — before they're announced or finalized. ● Strong written communication — comfortable working directly with an executive office to scope fast-turnaround, judgment-sensitive asks. Preferred / Nice to Have ● Prior exposure to financial/budget analysis, vendor cost reconciliation, or contingent-workforce (SOW) reporting. ● Familiarity with data engineering (SQL) and business intelligence tools (Looker, Power BI, Tableau) ● Prior work on an internal AI/automation platform team, or with a skills/plugin/tool registry and its certification pipeline. Roles & Responsibilities ● Build and maintain Claude Skills/Plugins for Strategy, Portfolio and Governance's recurring deliverables: CW/vendor cost-split analysis (SOW-level 4 by support tier, supplier reconciliation to source-of-truth totals), org-change one-pagers, MPR pre-read assembly, AI-adoption reporting, and hiring job description generation. ● Automate data pulls and reconciliation from source [contact hidden] — Workday (org/headcount), ServiceNow SPM/Master Portfolio, CW expense/SOW spreadsheets, and SharePoint/M365 — into standardized Strategy, Portfolio and Governance reporting formats, with built-in checks against known source totals. ● Develop and register lightweight MCP server integrations connecting Claude/Claude Code to the [contact hidden] the Strategy, Portfolio and Governance office relies on day to day (Workday, ServiceNow, SharePoint/M365, Slack, Jira/Confluence) via Cosmos AI Nexus MCP Hub. ● Certify and publish skills/plugins through the Cosmos Skills & Plugin Registry, meeting taxonomy, metadata, and CI compliance-validation requirements. ● Apply RBAC and data-handling guardrails correctly given the sensitivity of what flows through these tools — unannounced org changes, headcount/hiring plans, and vendor financials — so nothing is exposed outside a strict need-to-know. ● Work directly with team to translate ad hoc and recurring Strategy, Portfolio and Governance asks into repeatable automation, iterating quickly as reporting needs shift week to week. ● Maintain clear documentation for each skill/plugin (purpose, inputs/outputs, data sources, owner) and provide day-two support/maintenance once in production.
Recruiter contact
Unlock this recruiter's name, email and phone for a one-time $100. Payment is per role — you only pay for the introductions you actually want.
Want the full job-search package instead? Talk to us →
The recruiter's details are not public. Helen makes the introduction, checks your resume against the requirements first, and follows up on the reply.