AI Engineer

RoleBuild agents (planning and rollover) for a production LLM and deterministic Python tax pipeline, integrate them with PwC applications via REST APIs and agent tools (MCP), and deliver an agent-to-agent (A2A) pluggable agent for an enhanced graph-based orchestrator (LangGraph).Must-Have SkillsExperienced software engineer with 1+ years of Agentic Workflow development and enterprise API integration experiencePython, including:asyncioPydanticFastAPIpytest / pytest-asyncioBuilding LLM agents:Tool callingStructured outputsMulti-agent/sub-agent patternsCost and latency optimizationExperience with a mainstream LLM agent framework (e.g. Claude Agent SDK)MCP (Model Context Protocol): building and consuming tool serversLangGraph:Workflow nodesHuman-in-the-loop interruptsDurable checkpointingAgent-to-agent (A2A) protocolsEnterprise REST API integrations with OAuth2 / SSO (e.g. Azure AD)Redis / Redis StackDockerNice to HaveOpenTelemetry, Datadog, LangfuseAzure services (Blob Storage, Managed Search)LLM evaluations (LLM-as-judge)Spec-driven developmentTax or accounting domain experiencePrevious experience within the PwC ecosystem