Principal AI Platform Engineer

Capgemini • Atlanta, GA

5d • $10,000 - $140,000

About The Position

The role involves owning the knowledge layer for claims, payments, and health product lines. This includes managing the enterprise knowledge graph, the governed retrieval plane, context engineering, and the knowledge lifecycle for AI agents. It is a platform leadership role with a governance mandate, focused on making domain intelligence queryable and ensuring AI agent answers are grounded, relevant, and current.

Requirements

• Graph engineering depth: experience designing ontologies and graph schemas and operating production graph systems (e.g., Neo4j).
• RAG in production: experience building and optimizing retrieval pipelines, including embedding strategy, hybrid search, reranking, and retrieval evaluation.
• Context engineering fluency: ability to treat the context window as an engineered, budgeted artifact and build context assembly, compaction, or agent memory systems.
• Knowledge management for AI: curation, provenance, and versioning of enterprise knowledge for machine consumption.
• Regulated-data credentials: experience operating under financial-services data regulation or GDPR, with auditability designed from day one.

Responsibilities

• Design and manage the enterprise knowledge graph, including ontology and schema design, pipelines for building and refreshing it from production data, and graph retrieval as a platform capability.
• Optimize RAG (Retrieval-Augmented Generation) pipelines end-to-end, covering chunking and embedding strategy, hybrid and graph-augmented retrieval, reranking, and evaluation harnesses for grounding, relevance, and freshness.
• Establish context engineering as a platform discipline, defining what enters each agent's context window, in what order, and at what token budget, to be adopted by product lines.
• Manage the knowledge lifecycle for AI systems, from source documents to governed, versioned knowledge agents, including curation, provenance, freshness, and deprecation.
• Develop and govern the agent memory plane for writing and recalling episodic and precedent memory across product lines.
• Implement permission-aware retrieval, ensuring caller identity accompanies every query, including document-level access controls, multi-tenant isolation, and immutable audit trails.
• Build the data flywheel for model adaptation, using curated and governed production data for training and evaluation datasets for LLMs and SLMs.

Benefits

• Paid time off based on employee grade (Vacation: 12-25 days, depending on grade, Company paid holidays, Personal Days, Sick Leave)
• Medical, dental, and vision coverage
• Retirement savings plans (e.g., 401(k) in the U.S., RRSP in Canada)
• Life and disability insurance
• Employee assistance programs
• Other benefits as provided by local policy and eligibility

What This Job Offers

• Job Type: Full-time
• Career Level: Principal
• Education Level: No Education Listed