AI/ML Engineer (Graph Engineering and Generative AI)

Milestone 1 - Enhance the monitoring target state platform to perform AI based Quality Analysis / Quality Control on Issue Intake requests

Description:

• Leverage the existing monitoring target state platform to perform AI based quality analysis and quality control on BCM Issue Intake requests
• Apply standardized orchestration, prompt management, observability, and governance to improve consistency, accuracy, and auditability of intake quality assessments

Deliverables:

• Issue Intake QA/QC Workflows built using the existing orchestration and scheduling capabilities of the monitoring platform
• Quality Evaluation Prompts leveraging established prompt templates, prompt chaining, and prompt versioning for intake quality checks
• Intake Data Ingestion & Processing utilizing existing data connectors, storage, and processing patterns for unstructured request content
• QA/QC Execution Observability reusing platform logging, metrics, run status, error handling, retries, and audit trails
• Quality Scores & Outputs producing mathematical quality indicators and consumable results for BCM review and downstream reporting
• Documentation & BCM Enablement including intake QA/QC logic, operating guidance, and alignment to BCM control processes

Milestone 2 – Build a knowledge graph capability allowing BCMs to reference associated risks, issues, controls etc during Issue Intake, (plus other potential KG use cases)

Description:

• Build an AI driven knowledge graph capability that enables BCMs to automatically discover, reason over, and reference related risks, issues, controls, and policies during Issue Intake
• Leverage the monitoring platform’s AI orchestration, prompt management, observability, and governance capabilities to power intelligent context enrichment and decision support

Deliverables:

• AI Driven Knowledge Graph Model representing risks, issues, controls, policies, and relationships with semantic and contextual enrichment
• AI Based Entity Extraction & Linking leveraging GenAI to identify, classify, and relate entities from unstructured Issue Intake content
• Contextual AI Reasoning for Issue Intake enabling real time recommendations, relationship discovery, and impact analysis using KG augmented prompts
• KG Augmented Prompt Framework reusing existing prompt templates, prompt chaining, and prompt versioning to incorporate knowledge graph context
• Orchestrated AI Workflows leveraging existing scheduling, execution controls, and observability for KG population and inference
• Governance, Audit & Observability capturing AI decisions, entity relationships, prompt versions, and lineage for BCM compliance and control assurance.