Model Online / On-Site
Duration
Modules 6 modules

Course Objective & Scope

Standard single-shot RAG runs one search and passes whatever comes back to the LLM. Yet complex enterprise queries—such as comparing termination clauses across multiple contracts and cross-checking them against live SLA tables—require decomposing the question, querying multiple sources (vector stores, SQL, APIs) and reformulating the search when initial results fall short.

This course covers Adaptive RAG, Corrective RAG (CRAG) and Self-RAG patterns, stateful retrieval loops with LangGraph and LlamaIndex Workflows, and strict step, token and latency controls.

Who Should Attend?

  • AI engineers building research and analysis assistants where single-shot RAG falls short
  • Software engineers and architects combining document search with SQL, APIs and internal tools
  • Data scientists working on advanced RAG and agentic workflows
  • Prerequisite: solid grasp of RAG fundamentals, tool calling and Python development

Key Takeaways

  • Route simple queries to fast single-shot RAG and complex queries to multi-step flows (Adaptive RAG)
  • Decompose multi-hop questions into parallel or sequential sub-queries
  • Implement document grading and query rewriting loops (Corrective RAG) when retrieval is weak
  • Unify vector search, lexical search, SQL and internal APIs as agent tools
  • Bound latency and token spend using step limits, budgets and early-stopping criteria

Curriculum

From Linear Pipelines to Cyclic Agentic RAG
  • Where single-shot retrieval breaks: comparative queries, multi-hop synthesis and underspecified requests
  • Unconstrained ReAct loops versus explicit state-graph orchestration
Query Planning, Decomposition and Source Routing
  • Parallel sub-query generation and sequential multi-hop retrieval
  • Routing across vector stores, relational databases and operational APIs
Corrective RAG (CRAG) and Self-RAG Patterns
  • Grading retrieved chunks for relevance and triggering query rewrites on low confidence
  • Self-reflection loops checking hallucination and answer completeness before returning
Building Stateful Agentic RAG with LangGraph
  • Graph state schemas, nodes, conditional edges and parallel Map-Reduce retrieval branches
  • Streaming intermediate progress updates to the user interface
Controlling Cost, Latency and Runaway Loops
  • Preventing infinite rewrite loops with max iterations, token budgets and fallback exits
  • Using fast small models (SLMs) for grading and routing steps to keep latency low
Hands-On: Multi-Source Enterprise Research Assistant
  • Building a LangGraph Agentic RAG flow combining documents and structured data
  • Benchmarking accuracy vs latency against standard single-shot RAG

Other Courses in This Category

  • 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
  • Text-to-SQL and Conversational Analytics over Enterprise Databases

    Engineering Text-to-SQL and conversational analytics systems that translate natural-language business questions into accurate SQL queries, tables and charts over PostgreSQL, BigQuery, Snowflake and SQL Server.

    RAG & Enterprise Knowledge Systems Corporate
    Live Course
    6 modules
    View

Get in touch