AI-Augmented Automation Engineer

AI-Augmented Automation Engineer (L2): Addendum JD Read with the core CIMS Platform Engineer JD. The core requirements apply except where this addendum states otherwise. Role AI-Augmented Automation Engineer (L2): addendum to the core CIMS Platform Engineer JD. Replaces the working title “AI Forward-Deployed Engineer”. Seats (Year 1 / Year 3) 2 / 1 Location India-based. 24x7 shifts with handover overlap for the on-call rota. Core hours are the business day. Minimum experience 3+ years in operations or support engineering, plus 1+ year applying AI tools in production (proposed, pending approval). Reports to Technical Lead (L3) Purpose The client expects AI to speed up service delivery from day one. This role builds and runs the AI-assisted use cases for monitoring and the service disciplines: triage assistance, root-cause drafting, knowledge search and runbook drafting. Each use case is measured against a manual baseline and runs with human approval and an audit trail. Log sources include JBoss application logs, Oracle application tables, B2Bi and ITXA logs, and monitoring output. The role is an operations engineer who uses AI tools well, not a research or model-building role. Responsibilities Design and run AI-assisted use cases for monitoring and service disciplines, each with a manual baseline, a measured benefit, a control and an evidence method. Build assistance into the approved tools: GitHub Copilot for code and runbook work, and read-only access to logs and ServiceNow records through approved service accounts. Draft root-cause analyses from logs and incident records; check every claim against the evidence before it reaches the Problem Manager. Keep a prompt and skill library under version control, with peer review before any change. Run the evaluation log: accuracy, time saved, rejections and errors for each use case, reported monthly. Work with the monitoring and middleware engineers on the automation they need, under change control. Must-have skills Operations experience: incident handling, runbooks and shift work, to the level of the core JD (must-have). Applied AI tools in production work, with evidence: for example GitHub Copilot, prompt or skill libraries, or an approved assistant (must-have). Scripting and API use (Python or Bash), with error handling, timeouts and logging (must-have). ServiceNow or monitoring API integration through approved accounts (must-have). Evaluation and controls: measuring accuracy, human approval gates, data masking and audit logs (must-have). Log and incident reading for root-cause analysis, including Java stack traces and Linux logs (must-have). Git-based change: branches, reviews and tested releases (must-have). Nice to have Azure AI or machine-learning services awareness. Knowledge search and retrieval over runbooks and incident records. Nagios or LogicMonitor alert data analysis. Tools GitHub Copilot (McKesson-licensed); GitHub (Git); ServiceNow REST API; Nagios and LogicMonitor (read access); Python; Bash. How success is measured Each approved use case has a measured benefit against its manual baseline. Zero production changes made without human approval; every AI output that reaches a decision is logged. RCA drafting time falls against the baseline, with the accuracy checked by the reviewer. writer.editor.GO_TO_TOPwriter.editor.GO_TO_BOTTOM

Work Location: Remote