EXL—Support Analyst

Job description

Description: Technical Skills - knowledge of Hadoop ecosystem: Oozie Workflow Manager, Apache Hive, Apache Spark — job monitoring, resubmission, and basic troubleshooting. - Familiarity with Apache Airflow: DAG monitoring, task failure identification, and manual trigger/rerun procedures. - Experience with Managed File Transfer (MFT) tools — monitoring transfer jobs, identifying failures, and initiating reruns. - Working knowledge of Jenkins — pipeline monitoring, build status tracking, and failure alerting. - Understanding of AWS cloud services (S3, EMR, CloudWatch, EC2) in a data engineering context. - Experience using ServiceNow or similar ITSM platforms for ticket logging, triage, and workflow management. - Ability to read application/system logs and identify error patterns; comfortable with command-line interfaces. Operational & Soft Skills - 2–4 years of experience in IT operations or data platform support roles. - Ability to work in a rotational 24×7 shift environment including nights, weekends, and public holidays. - Strong attention to detail; disciplined in following runbooks, SOPs, and escalation protocols. - Clear and concise written communication for ticket documentation and handoff notes. - Collaborative team player; comfortable working with AI-assisted tools and adapting as automation evolves. - Ability to work calmly under pressure during high-severity (P1/P2) incident scenarios. Preferred Qualifications - ITIL Foundation certification (v3 or v4). - Exposure to data observability tools (Grafana, Prometheus, Cloud Watch or similar). - Basic scripting skills (Python/Bash) for log parsing and ad hoc data checks. - Experience in financial services or fintech data operations. Requirements: Graduate in Computer Science, B.Tech, B.E with 2+ years of hands-on data engineering experience. Responsibilities: Monitoring & Alerting - Proactively monitor data pipelines, batch jobs, Oozie workflows, Airflow DAGs, MFT transfers, and Jenkins jobs across all platform components. - Respond to automated alerts within SLA windows; validate, categorise, and action each alert using established runbooks. - Monitor Hadoop cluster health (HDFS, YARN, Spark, Hive); report anomalies and trigger job restarts/reruns per standard operating procedures. - Perform job monitoring and report cloud resource utilisation anomalies to the Lead. Incident Management & Triage - Register and confirm all incoming tickets in ServiceNow; acknowledge the issue logger within SLA (<15 minutes). - Categorise and prioritise tickets (P1–P4) using the Incident Triage Matrix; route to the correct resolution agent or support queue. - Perform initial troubleshooting using runbooks, RAG-enabled knowledge base, and FAQ repository; resolve known/routine issues without L2 escalation. - Escalate P1 incidents to L2 within 10 minutes with complete diagnostic context — logs, error messages, job details, and steps already taken. - Manage incident workflow through to closure; update ticket status, document resolution steps, and close with accurate categorisation. - Transition incidents to P2/P3 teams upon tiering changes; ensure smooth handover with full context preserved. Shift Handover & Documentation - Prepare comprehensive Daily Handoff Notes at end of shift: open incidents, in-progress issues, critical observations, and pending escalations. - Participate in shift-transition briefings with incoming analysts to ensure seamless knowledge transfer and continuity of monitoring. - Contribute to knowledge base enrichment by documenting new issue patterns, resolutions, and lessons learned to support RAG-based knowledge updates. Agentic AI Collaboration (HITL) - Operate as Human in the Loop (HITL) within the EXLdata.ai Agentic framework: review auto-triaged tickets, validate AI-suggested resolutions, and authorise automated remediation actions. - Provide feedback to the Agentic AI resolution and feedback agents to improve knowledge base accuracy and auto-resolution rates over time. - Utilise AI CoPilot suggestions to assist with complex ticket handling; escalate to L2 when AI and L1 tooling cannot resolve the issue.