Model Online / On-Site
Duration
Modules 5 modules

Course Objective & Scope

When starting an enterprise AI or semantic search initiative, one of the first architectural questions is whether pgvector inside existing PostgreSQL is enough, or when to adopt a dedicated vector database such as Milvus, Weaviate or Pinecone. This hands-on course puts all four technologies side by side on real code and datasets.

Participants implement schema setup, indexing, metadata filtering, hybrid (BM25 + dense) search, multi-tenancy and operational workflows across pgvector (HNSW, iterative scan, halfvec), distributed cloud-native Milvus, Weaviate and managed Pinecone Serverless. Teams leave with a working multi-engine PoC harness and an architectural decision matrix tailored to their scale and compliance constraints.

Who Should Attend?

  • Software and AI engineers building RAG pipelines, enterprise search and agent memory layers
  • Data, cloud and solution architects selecting the standard vector store for their organisation
  • Tech leads evaluating migrations between pgvector, self-hosted open-source engines and managed cloud services
  • Prerequisite: basic Python or backend development experience and familiarity with SQL/REST APIs

Key Takeaways

  • Tune PostgreSQL pgvector with HNSW indexes, halfvec quantization, iterative_scan and relational joins
  • Configure collections, partition keys, multi-vector (dense + sparse) search and index types in Milvus
  • Build class schemas, native BM25 + vector hybrid search and tenant-isolated shards in Weaviate
  • Implement namespace isolation, sparse-dense hybrid retrieval and metadata filtering on Pinecone Serverless
  • Select the right platform using a structured rubric covering scale, QPS, on-prem vs cloud governance, ACID joins and TCO

Curriculum

Relational Vector Search with PostgreSQL pgvector and pgvectorscale
  • vector, halfvec (FP16), bit and sparsevec types alongside distance operators (<=>, <->, <#>)
  • Building HNSW and IVFFlat indexes, tuning maintenance_work_mem and solving filtered search with hnsw.iterative_scan
  • Single-query RRF hybrid search combining PostgreSQL FTS (tsvector) with pgvector, plus pgvectorscale DiskANN
High-Scale Distributed Vector Search with Milvus
  • Milvus deployment modes (Lite, Standalone and Kubernetes Distributed with separate Query/Data/Index nodes)
  • Schemas, Partition-Key multi-tenancy, HNSW/IVF_SQ8/DiskANN indexes and configurable Consistency Levels
  • Multi-vector fields (Dense + BM25/Sparse) and server-side hybrid_search with RRF and WeightedRanker
Hybrid Search and Native Multi-Tenancy with Weaviate
  • Weaviate Collection schemas, inverted index settings and pluggable vectorizer/reranker modules
  • Built-in BM25 + Vector hybrid search using the alpha slider and Relative Score / Ranked Fusion
  • Native Multi-Tenancy with per-tenant shards and HOT/COLD/FROZEN lifecycle offloading
Managed Vector Infrastructure with Pinecone Serverless
  • Pinecone Serverless storage-compute separation, index setup and Namespace isolation
  • Batched upserts, selective metadata filters and Sparse-Dense hybrid vector queries
  • Optimising Read/Write Units, point-in-time backups and enterprise security controls
Architectural Decision Matrix and Side-by-Side PoC Lab
  • Comparing pgvector, Milvus, Weaviate and Pinecone across scale, latency, ops burden, on-prem needs and TCO
  • Hands-on lab: running identical hybrid retrieval workloads across engines and evaluating results

Other Courses in This Category

  • Database Performance Tuning and SQL Query Optimization

    Cost-Based Optimizer internals, reading Execution Plans, SARGability, B-Tree/Covering/Filtered index design, join algorithms, parameter sniffing, lock/isolation analysis and partitioning — for developers and DBAs.

    Databases Corporate
    Live Course
    5 modules
    View
  • Oracle PL/SQL Database Programming and Performance

    PL/SQL block structure, cursors, modular Package design with procedures and functions, enterprise exception handling, triggers, dynamic SQL and high-volume processing with FORALL and BULK COLLECT.

    Databases Corporate
    Live Course
    5 modules
    View
  • Relational Data Modelling, SQL Normalization and Schema Optimization

    Conceptual/logical/physical data modelling, 1NF–3NF/BCNF normalization rules, preventing data anomalies, controlled denormalization, natural vs surrogate keys, constraints and zero-downtime schema evolution.

    Databases Corporate
    Live Course
    5 modules
    View
  • SQL Server Database Querying Fundamentals SQL Querying Fundamentals

    Foundational SQL Server course covering relational database concepts, data modeling, normalization, indexing, database objects (Tables, Views, Stored Procedures), and JOIN queries.

    Databases Corporate
    Live Course
    8 modules
    View
  • SQL Server Database Querying Training SQL Server Querying

    Comprehensive SQL Server and T-SQL querying course covering SSMS, SELECT workflows, multi-table JOINs (INNER, OUTER, CROSS, SELF), filtering and pagination, data types, and built-in Scalar, Aggregate, and Window functions.

    Databases Corporate
    Live Course
    7 modules
    View
  • Big Data Querying and Analytics with Apache Spark SQL

    Apache Spark SQL and the DataFrame API, Catalyst Optimizer and Adaptive Query Execution (AQE), Parquet/Delta Lake/Iceberg table formats, join strategies and fixing data skew — for teams building big-data analytics.

    Databases Corporate
    Live Course
    5 modules
    View

Get in touch