Job Description

Experience: 6 Years +

**Key Responsibilities**

Design and develop a production-grade Multi Agent Runtime that manages agent lifecycle, state persistence, audit logging, and orchestration of specialized diagnostic agents
Build and optimize the Principal Agent that dynamically schedules tasks and aggregates results from downstream agents (Performance, Topology, Alarm, Ticket, CX, and Event Scout) using the Model Context Protocol (MCP)
Implement the RCA Inspector reasoning engine that analyzes multi-source incident context against structural playbooks and historical data to produce ranked, explainable root cause hypotheses
Maintain stream-to-stream API contracts and integrations with neighboring systems, including Data Stream (Pub/Sub anomaly consumption), Knowledge Stream (semantic graph queries), and Experience Stream (WebSocket/SSE operator dashboards)
Build secure, human-in-the-loop integrations for the Remediation Agent to interface with backend APIs, enabling controlled actions such as traffic reroutes, ticket creation, and partner notifications under strict approval gates
Ensure all agent decisions, intermediate steps, and reasoning are logged and reconstructable for full auditability and explainability