Skip to content
New this month 24 fresh C++, C#, F#, JavaScript, TypeScript and Lua courses just landed. Browse new releases Use code WELCOME10 for 10% off your first order · 14-day refund
WisdomCharms — home
Cart 0 items in cart
AI & Machine Learning Intermediate

Vector Stores in Practice: pgvector, Qdrant and Friends

Pick, index and operate a vector store without turning your search into a science project.

Rated 4.0 out of 5 from 2 reviews 9 students

PR Created by Priya Raman

  • Last updated August 2026
  • English
  • 6h 35m of material
  • 49 lessons

What you will learn

8 concrete outcomes

Every bullet below is something you will have built, shipped or be able to explain by the time you finish the last lesson.

  • Explain flat, inverted file and HNSW indexes and the trade each one makes
  • Measure real recall against exact search instead of trusting a claim
  • Tune index parameters knowing what each does to memory, build time and recall
  • Compare pgvector, Qdrant and a third store on identical data and queries
  • Understand pre-filtering and post-filtering and why the difference matters
  • Reindex safely when you change embedding model or dimension
  • Monitor retrieval quality in production rather than only at launch
  • Build a cost model for each option at small, medium and large corpus sizes

Course curriculum

6 modules · 49 lessons · 6h 35m of material

8 lessons running 1h in total. Each lesson ships with the finished source files and a short written recap, so you can follow along in your own editor and skim the module again later.

Lesson-by-lesson titles, code downloads and exercises live inside the course library you get access to straight after checkout.

8 lessons running 1h 6m in total. Each lesson ships with the finished source files and a short written recap, so you can follow along in your own editor and skim the module again later.

Lesson-by-lesson titles, code downloads and exercises live inside the course library you get access to straight after checkout.

8 lessons running 1h 10m in total. Each lesson ships with the finished source files and a short written recap, so you can follow along in your own editor and skim the module again later.

Lesson-by-lesson titles, code downloads and exercises live inside the course library you get access to straight after checkout.

8 lessons running 1h 10m in total. Each lesson ships with the finished source files and a short written recap, so you can follow along in your own editor and skim the module again later.

Lesson-by-lesson titles, code downloads and exercises live inside the course library you get access to straight after checkout.

8 lessons running 1h 6m in total. Each lesson ships with the finished source files and a short written recap, so you can follow along in your own editor and skim the module again later.

Lesson-by-lesson titles, code downloads and exercises live inside the course library you get access to straight after checkout.

9 lessons running 1h 3m in total. Each lesson ships with the finished source files and a short written recap, so you can follow along in your own editor and skim the module again later.

Lesson-by-lesson titles, code downloads and exercises live inside the course library you get access to straight after checkout.

6 modules · 49 lessons

6h 35m total length

Requirements

Short list, and deliberately so. If you meet these you can start today.

  • Comfortable with Python, SQL and the command line
  • Docker available for running the stores locally
  • Some prior exposure to embeddings and similarity search

About this course

Choosing a vector store is one of those decisions teams make in twenty minutes and live with for two years. This course gives the decision the time it deserves, and then teaches you to operate whatever you chose.

You begin with the index structures themselves, because the marketing pages will not explain them and the behaviour you care about comes directly from them. Flat search and when it is genuinely the right answer. Inverted file indexes and the list count trade. Hierarchical navigable small world graphs, the parameters that govern build time, memory and recall, and what actually happens to results when you tune each one.

Then you run the same workload on PostgreSQL with pgvector, on Qdrant and on a third store, using identical data and identical queries. You compare ingestion speed, memory footprint, filtered search behaviour, update cost and operational burden. The filtering comparison alone changes most people's choice, because pre-filtering and post-filtering produce very different results and very different latency.

The final third is operations: bulk loading a large corpus, reindexing when you change embedding model, backups, monitoring recall in production, sharding when a single node is no longer enough, and the cost model for each option at three corpus sizes. You finish able to defend your choice with numbers.

Topics covered

Frequently asked questions

Still unsure about something? Write to misteryjj100@gmail.com and a human answers, usually the same working day.

It will tell you how to decide. For most teams under a few million vectors pgvector wins on operational simplicity, and the course shows exactly where that stops being true.

No. A one million vector corpus is generated by a provided script, which is enough to make the differences between index types clearly visible on a laptop.

Managed offerings are compared on cost and operational burden, but every exercise runs locally through Docker so nothing requires a paid account.

Checkout is handled on our provider's secure payment page. The moment your payment clears we email your personal access link and access code to the address you used at checkout, and the same link appears in your account library. There is nothing to install and nothing to wait for.

Email misteryjj100@gmail.com within 14 days of your purchase, quote your order number, and we refund the full amount to your original payment method. No form to fill in and no questions about how much of the course you watched.

What students say

Reviews are written by people who bought this course. We publish the critical ones too.

4.0

Rated 4.0 out of 5

Course rating · 2 reviews

Rating distribution

  • 5 stars 50%
  • 4 stars 0%
  • 3 stars 50%
  • 2 stars 0%
  • 1 star 0%
  • AM

    Anton Melnyk

    Backend engineer

    Feb 2026
    Rated 5.0 out of 5

    Measuring recall honestly

    Building a ground-truth set with exact search first, then measuring what your approximate index actually returns against it, is the step everybody skips. This course makes you do it before you are allowed an opinion about which store is better.
  • LK

    Line Kristensen

    Search engineer

    Aug 2025
    Rated 3.0 out of 5

    Filtered search deserved more

    Index structures and the recall methodology are both strong. But filtered vector search, where a metadata filter interacts badly with an approximate index and recall quietly collapses without any error being raised, gets a short section that names the problem and then moves on. That is the hardest thing about operating one of these in production and it needed a chapter of its own.

Your instructor

PR

Priya Raman

Applied AI and machine-learning engineer

  • 236 students taught
  • 13 courses published
  • 4.3 instructor rating
  • Machine learning
  • RAG
  • Fine-tuning
  • LLM agents

Priya builds language-model features for a document-heavy SaaS product, which means she has taken retrieval and fine-tuning from a promising notebook to something on call at three in the morning. She teaches the mathematics only where it changes a decision you are about to make, and spends the rest of the time on data quality, evaluation and the cost of an agent that loops. Her courses run on a laptop and a modest API budget, so nobody has to rent a cluster to follow along. She publishes reproducible notebooks alongside every module.

See every course by Priya Raman
$45 USD

One-time payment · lifetime access

The WisdomCharms dispatch

One useful email a week. No fluff, no spam.

New course releases, discount codes before anyone else, and a short, practical breakdown of one technique — a prompt pattern, a C++ idiom, a TypeScript trick — that you can use the same day.

  • Subscriber-only launch pricing
  • Unsubscribe in one click
  • We never sell your address

By subscribing you agree to our Privacy Policy. Questions? Write to misteryjj100@gmail.com.