AI Application Developer

Design and develop backend services for the AI-based Application.

• Implement operator-facing service flows supporting dashboards, exception workspaces, context inspection, and learning governance workflows.

• Develop transactional APIs for operator commands such as acknowledge, resolve, escalate, annotate, and case updates.

• Implement real-time event streaming for exception arrival, agent progress, queue updates, and re-evaluation events.

• Build and enhance EDI/X12 exception processing components including EDI Parser, exception normalization, and internal exception model mapping.

• Implement deterministic exception resolution flows including L1 reference-data lookups and L2 policy-based routing.

• Develop Agent Orchestrator workflows for L3 agentic resolution when deterministic rules cannot resolve exceptions.

• Integrate bounded resolution tools such as drug lookup, pricing check, serial verification, DEA check, and document lookup.

• Build integration with Graph RAG and Document RAG capabilities for contextual retrieval and decision support.

• Develop services such as Exception Service, Context Query Service, Learning Service, and Decision Trace Emitter.

• Implement decision trace capture for deterministic and agentic decisions to support auditability, explainability, and learning loops.

• Build ingestion and retrieval integrations using authoritative external sources such as drug registries, DEA registry, pricing data, serialization data, policies, documentation, and regulatory guidance.

• Implement feedback ingestion, pattern mining support, candidate rule synthesis support, routing optimization support, validation sandbox integration, and promotion workflows.

• Integrate application services with Azure Databricks, Unity Catalog, Mosaic AI, vector stores, graph stores, object stores, and model-serving services.

• Build REST, HTTPS, WebSocket, JSON-RPC, event-driven, and asynchronous interfaces as required.

• Implement secure access patterns using authentication, authorization, role-based access control, data governance, and audit controls.

• Ensure idempotent processing, metadata tracking, lineage, traceability, and consistency across persistence layers.

• Collaborate with QA, DevOps/MLOps, data engineering, architecture, security, and business teams.

• Participate in code reviews, design discussions, troubleshooting, performance tuning, and production readiness activities.