AI Engineer - Lead

1. Engineer knowledge ingestion pipelines

Build and operate the technical pipelines used to ingest knowledge into Twin.

Responsibilities include:

·         Implement ingestion workflows for documents, procedures, FAQs, incident knowledge, change knowledge, infrastructure knowledge and operational standards. 

·         Extract content from structured and unstructured sources. 

·         Normalize formats before ingestion. 

·         Apply metadata enrichment. 

·         Manage document versioning and lifecycle metadata. 

·         Prepare content for RAG indexing. 

·         Prepare content for graph-based representation when relevant. 

·         Detect ingestion errors, incomplete documents and unsupported formats. 

·         Implement automated quality checks before exposure to Twin agents. 

2. Structure knowledge for RAG and GraphRAG

Transform raw knowledge into AI-usable knowledge.

Responsibilities include:

·         Define and apply chunking strategies. 

·         Implement semantic tagging. 

·         Add metadata such as owner, source system, validity date, confidentiality level, domain, franchise, CI, application, service and operational scope. 

·         Optimize retrieval quality through indexing strategy. 

·         Support hybrid search, semantic search and graph-assisted retrieval. 

·         Link knowledge chunks to entities in the knowledge graph. 

·         Maintain relationships between applications, infrastructure components, procedures, incidents, changes and services. 

·         Improve grounding and source citation capabilities. 

3. Build and maintain knowledge graph content

Support the creation and maintenance of graph-based knowledge used by Twin.

Responsibilities include:

·         Model key IT production entities and relationships. 

·         Populate graph structures from documentation, CMDB, inventories and operational sources. 

·         Support entity resolution and deduplication. 

·         Detect inconsistent relationships. 

·         Contribute to graph quality checks. 

·         Support GraphRAG use cases. 

·         Validate graph traversal results used by agents. 

·         Work with Twin Core and graph specialists on ontology, schema and graph evolution. 

4. Implement knowledge quality controls

Develop controls to ensure that Twin consumes trusted and usable knowledge.

Responsibilities include:

·         Implement automated checks for missing metadata. 

·         Detect obsolete documents. 

·         Detect duplicated or conflicting content. 

·         Flag documents without owner or validation date. 

·         Track source freshness. 

·         Monitor knowledge coverage by domain or franchise. 

·         Detect knowledge pollution or drift. 

·         Implement rules to prevent unvalidated content from being exposed to critical agents. 

·         Produce quality reports for knowledge owners and Twin franchises. 

5. Support source certification and exposure rules

Engineer the technical mechanisms that control which knowledge sources can be used by Twin agents.

Responsibilities include:

·         Maintain a technical registry of knowledge so