GENERAL JOB DESCRIPTION
Hands-on L2/L3 application support and reliability engineers covering Windows (IIS, .NET), Linux (Tomcat, Java, Mirth Connect), Kubernetes microservices, and GPU-accelerated AI inference endpoints. Ensure 24x7 stability, SLA-driven incident resolution and proactive maintenance across heterogeneous stacks.
DUTIES & RESPONSIBILITIES
- Triage, investigate and resolve L2/L3 incidents within agreed SLAs.
- Perform deep-dive log analysis across IIS, Tomcat, Mirth Connect, MS SQL, PostgreSQL, Kubernetes pods and inference services.
- Execute approved changes, releases and database / cluster housekeeping.
- Monitor health and capacity of application, integration, DB and GPU tiers.
- Participate in 24x7 on-call rotation.
- Support DR drills and failover / failback execution.
- Document RCAs and maintain the known-error knowledge base.
- Engage Level-4 vendor support (database, middleware, GPU and inference vendors) when required.
SKILLS & ABILITIES
- Strong Windows Server and Linux administration.
- IIS, Tomcat and Mirth Connect troubleshooting.
- MS SQL Server and PostgreSQL query analysis and basic performance tuning.
- Kubernetes operations: kubectl, Helm, logs, events, network policies.
- Familiarity with GPU/inference services (NVIDIA Triton, vLLM) and NVIDIA DCGM is a strong plus.
- Scripting in PowerShell, Bash and/or Python.
- HL7 v2.x message debugging.
- ITSM tools (ServiceNow / Jira SM).
POTENTIAL BACKGROUND
- Bachelor's in CS / Engineering or related.
- 4+ years in production application support for healthcare, public-sector or cloud/AI platforms.
- Experience with enterprise healthcare applications is a strong plus.
- Exposure to Kubernetes operations and AI inference workloads is highly desirable.
PREFERRED TOOLS / SOFT SKILLS
Preferred tools:
- IIS, Tomcat, Mirth Connect
- MS SQL Server, PostgreSQL
- Kubernetes (kubectl, Helm)
- NVIDIA Triton, vLLM, NVIDIA DCGM
- PowerShell, Bash, Python
- Grafana, Zabbix, Azure Monitor
- ServiceNow, Jira Service Management
Soft skills:
- Ownership mindset
- Calm troubleshooter
- Disciplined documentation
- Customer-focused
