Principal Network Automation Engineer

Business Introduction GSK remains committed to achieving bold commercial ambitions for the future. By 2031, we aim to deliver £40 billion in annual sales, leveraging our existing strong performance momentum to significantly increase our positive impact on the health of billions of patients globally. Our Ahead Together strategy is centred on early intervention to prevent and alter the course of disease, thereby protecting people and supporting healthcare systems. Our diverse portfolio consists of vaccines, specialty medicines, and general medicines. Through continuous innovation and a dedicated focus on scientific and technical excellence, we strive to develop and launch new, groundbreaking treatments that address critical health challenges. Job Purpose We're looking for a Network Automation AI Engineer to design, build, and operate AI agents that automate network operations, ITSM, and lifecycle management across a large enterprise network estate. You'll combine traditional network automation (Cisco, Itential, ServiceNow) with modern agentic AI (LLM orchestration, MCP tool integration, RAG) to build production-grade, safety-gated automation. Responsibilities • Design and build AI agents (using frameworks like Google ADK, LangChain, or similar) that interact with network infrastructure via Model Context Protocol (MCP) servers or custom tool integrations • Write and iterate on system prompts/instructions that ground LLM behavior in real tool schemas, enforce anti-hallucination discipline, and avoid unsafe automation • Integrate agents with ITSM platforms (ServiceNow) for incident, change, and CMDB data, and with network orchestration platforms (Itential) for workflow execution • Build human-in-the-loop approval workflows for any agent capability that can modify production network state (e.g., plan → approve/reject → apply patterns), with strict identity/authorization guardrails • Develop and maintain backend services (Python, FastAPI) exposing agents via REST/A2A/JSON-RPC protocols • Containerize and deploy agents (Docker, Cloud Run, CI/CD via GitHub Actions) with per-agent configuration, environment isolation, and secrets management • Instrument agents with observability (OpenTelemetry, structured logging, Cloud Logging) for tool I/O, session traces, and audit trails • Collaborate with network engineering teams to translate operational runbooks (EOL/EOS tracking, device lifecycle, LAN/WAN/wireless health checks, config change validation) into agent capabilities • Own RBAC/token propagation and security review for agent-to-tool and agent-to-user authentication • Test agents against live/staging MCP servers, validate tool schemas, and iterate on prompt engineering based on real session traces • Document architecture decisions, known gotchas, and operational runbooks for the agent fleet Required Qualifications • Overall 10 Years and strong Python skills; experience building async services (FastAPI, asyncio) • Hands-on experience with LLM application development: prompt engineering, tool/function calling, RAG, multi-turn session/state management • Familiarity with agent frameworks (Google ADK, LangChain, LangGraph, Semantic Kernel, or similar) and agent-to-agent/tool protocols (MCP, A2A) • Networking fundamentals: Cisco IOS/IOS-XE, LAN/WAN/wireless architecture, routing, switching, firewalls • Experience with ITSM tooling (ServiceNow) and network orchestration platforms (Itential, NSO, or similar) • Docker containerization and CI/CD pipeline experience (GitHub Actions or equivalent) • Cloud platform experience (GCP preferred: Vertex AI, Cloud Run, BigQuery, Secret Manager) • Strong understanding of security practices for automation that can modify production infrastructure (least privilege, approval gating, audit lo