Director, Analytics

Description & Requirements

• Lead the Data & Intelligence Engineering discipline across the AAI practice — setting technical direction, quality standards, and delivery patterns for the Layer 1 Data Foundation that supports delivery.
• Provide coaching, career development, and performance feedback to Data & Intelligence Engineers.
• Partner with peer leaders of the AI Solutions Architects and Intelligence Designers to define solution architecture, hand-offs across the three logic layers (Data, Reasoning, Validation), and shared standards for MCP, evals, and HITL.
• Contribute to the AAI Center of Excellence (Hub) — selecting frontier tooling, codifying standards (Zero-ETL, vector DBs, knowledge graphs, MCP), and ensuring cross-team knowledge sharing.
• Represent the practice on AI Ethics, model governance, data sovereignty, and responsible AI guardrails, in alignment with the AAI Steering Committee and Client Zero principles.
• Architect and oversee the Data Foundation layer of agentic systems: vector stores (Pinecone, Weaviate, Milvus), knowledge graphs (Neo4j), semantic layers, and Zero-ETL / real-time streaming pipelines.
• Thorough understanding of data modeling, relational, and non-relational principles; expert level SQL
• Expertise in Python data engineering packages and patterns, including Python and PySpark
• Familiarity and experience with the different data engineering platforms commonly used such as Fabric, Databricks, Snowflake, Synapse
• Solid leadership-level fluency with cloud fundamentals (Azure preferred, GCP, AWS) and modern data architecture (lakehouse, data contracts, governance, lineage)
• Expected to actively collaborate with clients and team members to meet project objectives and timelines; Strong communication skills (verbal and written) for executive, client, and team audiences.
• Experience with building, testing and deploying data pipelines, ML/AI models; understanding of GenAI/Agentic AI use cases (experience strongly preferred)
• Design and govern MCP (Model Context Protocol) connectors that securely expose enterprise sources (Jira, CRM, ERP, lakehouse) as tools consumable by agents and frontier models.
• Set standards for Graph-RAG, RAG optimization, chunking strategy, and retrieval evaluation to minimize hallucinations and ensure high-quality context for downstream LLM components.