Model Online / On-Site
Duration
Modules 6 modules

Course Objective & Scope

The biggest blocker to deploying an enterprise knowledge assistant is the risk of retrieving and exposing a document—such as an executive memo, payroll sheet or another client's contract—that the asking user is not authorised to see. Likewise, vector search without metadata filtering routinely mixes outdated policies with current ones.

This course covers enterprise metadata schema design, natural-language filter extraction (self-querying), and permission-aware (security-trimmed) retrieval architectures that enforce RBAC, ABAC and source-system ACLs (SharePoint, Google Drive, Confluence, DMS) at query time.

Who Should Attend?

  • AI and backend engineers taking enterprise RAG and internal knowledge assistants to production
  • Solution architects responsible for data security, IAM integration and multi-tenant isolation
  • Platform teams connecting SharePoint, Confluence, DMS and internal repositories to AI services
  • Prerequisite: familiarity with RAG architectures, vector databases and authentication/authorisation (JWT/OAuth/RBAC)

Key Takeaways

  • Design an enterprise metadata taxonomy that improves retrieval precision and recency handling
  • Build automated metadata extraction and tagging pipelines during document ingestion
  • Implement self-querying retrievers that translate natural-language constraints into structured filters
  • Enforce query-time security trimming tied to verified user identity (OIDC/Entra ID) and ACLs
  • Handle permission revocation sync and run adversarial access-control leak tests

Curriculum

Metadata Architecture and Taxonomy for Enterprise RAG
  • Source, owner, department, effective date, version, classification level and entity tags
  • Automated metadata extraction at ingestion using deterministic rules and structured LLM calls
Dynamic Filtering and Self-Querying Retrievers
  • Translating natural-language date and category constraints into vector database filter expressions
  • Filter relaxation patterns to avoid zero-result queries when inferred filters are too strict
Permission-Aware Retrieval: RBAC, ABAC and ReBAC
  • Why access control must happen in the retrieval/storage layer, never via system prompt instructions
  • User/group/role filtering, OpenFGA/Zanzibar relationship-based access control and tenant isolation
Source ACL Synchronisation and Revocation Handling
  • Mapping SharePoint, Google Drive and Confluence allow/deny ACLs into the search index
  • Trade-offs between index-time (early binding) and query-time (late binding) permission checks
Preventing Leaks in Derived Artefacts (Caches, Summaries, Memory)
  • Cross-user leakage risks in semantic caches, conversation memory and pre-computed summaries
  • Permission-scoped cache keys and compliance audit logging
Hands-On: Building a Security-Trimmed Enterprise Knowledge Assistant
  • Implementing permission-aware retrieval across multi-role user personas
  • Running red-team verification tests for unauthorized document and inference leakage

Other Courses in This Category

  • Agentic RAG: Multi-Step, Corrective and Adaptive Retrieval Architectures

    Building Agentic RAG systems with LangGraph that plan sub-queries, route across heterogeneous data sources, grade retrieved evidence, self-correct queries (CRAG/Self-RAG) and bound latency.

    RAG & Enterprise Knowledge Systems Corporate
    Live Course
    6 modules
    View
  • GraphRAG and Knowledge-Graph-Enhanced Retrieval Systems

    Combining knowledge graphs (Neo4j, property graphs), entity-relationship extraction, community summaries (Microsoft GraphRAG) and vector search to answer multi-hop and corpus-wide enterprise queries.

    RAG & Enterprise Knowledge Systems Corporate
    Live Course
    6 modules
    View
  • RAG Evaluation and Quality Assurance: Golden Datasets, LLM-as-a-Judge and CI/CD

    Replacing ”vibe checks” in RAG with systematic evaluation: golden dataset curation, synthetic test generation, RAGAS/DeepEval/Promptfoo pipelines, CI/CD quality gates and production monitoring.

    RAG & Enterprise Knowledge Systems Corporate
    Live Course
    6 modules
    View
  • Retrieval and Generation Metrics for RAG Systems

    Hands-on programme covering information retrieval metrics (Recall@k, Precision@k, MRR, NDCG@k) and generation metrics (Faithfulness, Answer Relevance, Context Precision, Citation F1) with statistical testing.

    RAG & Enterprise Knowledge Systems Corporate
    Live Course
    6 modules
    View
  • Production RAG Architecture: Scaling, Freshness, Security and Operations

    Scaling RAG from prototype to production: event-driven incremental ingestion, zero-downtime re-indexing, multi-tier caching, security controls, latency SLAs and LLMOps observability.

    RAG & Enterprise Knowledge Systems Corporate
    Live Course
    6 modules
    View
  • Building an Enterprise Knowledge Assistant: Architecture and Implementation

    End-to-end engineering programme for building an enterprise knowledge assistant across SharePoint, Confluence, Jira and internal docs with multi-turn memory, citations, permissions and feedback loops.

    RAG & Enterprise Knowledge Systems Corporate
    Live Course
    6 modules
    View

Get in touch