-
Module 1: AI in the context of PostgreSQL
11 Lessons-
StartModule introduction
-
StartAI features in PostgreSQL
-
StartWhat the database is responsible for
-
StartWhat the database is not responsible for
-
StartWhat the model service is responsible for
-
StartExtensions vs PostgreSQL core
-
StartMisconceptions to avoid
-
StartPositioning PostgreSQL correctly
-
StartLab 1
-
StartLab 1 video walkthroughs
-
StartQuiz 1
-
-
Module 2: Vector data and embeddings fundamentals
12 Lessons-
StartModule introduction
-
StartWhat is an AI model?
-
StartWhat embeddings are and why they exist
-
StartHow vector similarity differs from relational comparison
-
StartTypical AI scenarios that involve vectors
-
StartWhy vectors work well for these scenarios
-
StartWhen vectors are inappropriate
-
StartWhere embeddings models are hosted
-
StartCommon embeddings models
-
StartLab 2
-
StartLab 2 video walkthroughs
-
StartQuiz 2
-
-
Module 3: Vector data types in PostgreSQL
10 Lessons-
StartModule introduction
-
StartVector data type basics
-
Startpgvector extension basics
-
StartDeclaring vector columns and variables
-
StartHow vector values are represented
-
StartPractical table pattern for embeddings
-
StartDimensionality is enforced
-
StartDimensionality changes are migration events
-
StartStorage characteristics and limitations of pgvector
-
StartQuiz 3
-
-
Module 4: Querying vector data
7 Lessons -
Module 5: Vector indexing and performance
8 Lessons -
Module 6: Integrating External AI Services
11 Lessons-
StartModule introduction
-
StartWhy PostgreSQL might need an external AI service
-
StartWhat REST means in practice
-
StartHTTP requests, responses, headers, and status codes
-
StartJSON request and response payloads
-
StartEndpoints, hosts, ports, and network reachability
-
StartAuthentication and secrets
-
StartTimeouts, throtting, quotas, and cost
-
StartError handling, retries, and idempotency
-
StartOperational risks of outbound calls from the database
-
StartQuiz 6
-
-
Module 7: Calling AI services from PostgreSQL
13 Lessons-
StartModule introduction
-
StartWhere should AI model calls be made ?
-
StartSynchronous HTTP and asynchronous workflows
-
StartHTTP Options from PostgreSQL
-
StartIntroducing pgsql-http
-
StartRunning AI models locally with Ollama
-
StartCalling Ollama from PostgreSQL with pgsql-http
-
StartProcessing responses and storing embeddings
-
StartInvoking generative and chat models
-
StartHandling timeouts, failed requests, and retries
-
StartTracking embedding model and version information
-
StartCalling online models from PostgreSQL
-
StartQuiz 7
-
-
Module 8: Creating AI Pipelines
13 Lessons-
StartModule introduction
-
StartWhy AI enrichment is usually background work
-
StartIdentifying rows that require processing
-
StartTracking embedding state and model versions
-
StartEfficiently finding pending work
-
StartBackground execution options
-
StartTwo ways to execute the model call
-
StartDurable state and transaction boundaries
-
StartHandling retries, concurrency, and idempotency
-
StartRe-embedding when content or models change
-
StartMonitoring backlog, failures, and throughput
-
StartWhere a workflow engine might be appropriate
-
StartQuiz 8
-
-
Module 9: Implementing pipelines on Windows with pgsql-http
9 Lessons-
StartModule introduction
-
StartWindows lab architecture
-
StartPreparing the Windows extensions
-
StartPreparing rows for background embedding
-
StartCalling Ollama in batches with pgsql-http
-
StartScheduling background processing with pg_cron
-
StartOne processing cycle
-
StartObserving and recovering the Windows pipeline
-
StartQuiz 9
-
-
Module 10: Implementing pipelines on Unix with pg_net
9 Lessons-
StartModule introduction
-
StartIntroducing pg_net
-
StartUnderstanding the pg_net request lifecycle
-
StartCalling Ollama from PostgreSQL with pg_net
-
StartDispatching batches asynchronously with pg_net
-
StartTracking requests and processing responses
-
StartRetry and recovery with pg_net
-
StartChoosing between pg_net and pgsql-http
-
StartQuiz 10
-
-
Module 13: When NOT to use AI features in PostgreSQL
7 Lessons
Using AI Features in PostgreSQL (In development - coming soon)
Store, retrieve, and integrate data for AI systems using PostgreSQL
Course Summary
Do you want to start working with AI-related features in PostgreSQL?
As these features evolve, the focus of this course is on design and architecture decisions that remain valid, even as individual features, syntax, and external AI services evolve. But we will show you how to achieve great outcomes right now. It does not aim to turn PostgreSQL into an AI platform, nor does it attempt to teach model training, data science, or vendor-specific AI tooling.
This course is about using PostgreSQL to store, retrieve, and integrate data for AI systems, in a way that is reliable, understandable, and operationally sound. We assume you already understand PostgreSQL fundamentals and are comfortable working with PostgreSQL and database design.
— Are you responsible for deciding where AI-related logic should live, and where it should not?
— Do you need to support vector search, retrieval, or AI integration without turning the database into an experimental system?
— Have you been asked whether PostgreSQL should be used for embeddings, RAG patterns, or AI orchestration, and wanted a clear answer that you can explain?
If that matches the kind of responsibility you have, you’re in the right place. You'll get clear explanations, hands-on exercises, and short quizzes to make sure you not only understand the concepts but can confidently apply them in real PostgreSQL environments. And while the practical exercises are optional, we strongly encourage completing them. They are designed to be straightforward and focused on realistic scenarios, rather than tooling complexity. Setup is easy.
