New: Foundations of Generative AI for Business — Now Available

Level Up With
Data Science Dojo

Comprehensive, hands-on courses to master AI, LLMs & Data Science.
Start free — from first lesson to production-ready.

Start Learning Free

3 Free BundlesStart today, no card

Certificate EarnedIndustry-recognized

12,000+ LearnersGlobal community

Trusted by professionals at

Microsoft Apple Orcale Nike NAES Google KPF Microsoft Apple Orcale Nike NAES Google KPF

Professionals Trained

Companies Represented

M+

Community Members

Average SwitchUp Rating

Explore by Category

Find your next learning path

Browse our curated course collections — from agentic AI to analytics.

Inside the Academy

Walk through a complete course experience

From bootcamp overview to your final AI project — explore every step of how you'll learn inside DSD Academy.

› › › › ›

Learning Path

Structured from day one

Courses are sequenced so every concept builds on the last — no jumping around.

Video lessons Notebooks Activities Quizzes Final project

Foundations

Video · Notebook · Quiz

Core Concepts

Video · Activity · Notebook

Applied Skills

Video · Notebook · Quiz

Advanced Topics

Video · Activity · Notebook

Expert Practice

Video · Quiz · Notebook

Capstone Project

AI Agent · Deploy · Certificate

CAPSTONE

AI Fundamentals

12% complete

Search by lesson title

Vector Databases 0/5

What are Vectors?VIDEO · 8 MIN

Vector DatabasesVIDEO · 12 MIN

HNSW GraphACTIVITY

Cosine Similarity SearchNOTEBOOK

Chapter QuizQUIZ · 5 Q

Embeddings & Search0/4

RAG Pipelines0/6

Production & Scaling0/5

Capstone Project0/3

Vector Databases

0 Discussions⤢

Vector Databases

Embeddings · Similarity Search · Indexing

4:12 / 11:48 CC ⚙

HNSW Graph Navigation

0 Discussions

ACTIVITY

HNSW Graph Navigation

Watch how Hierarchical Navigable Small World graphs find the nearest neighbor using greedy search — hopping from node to node until reaching the query.

Entry Current Visited Query

Distance—

Current Node—

Steps Taken0

Neighbors Checked0

Cosine Similarity Search

0 Discussions

Cosine Similarity Search

Rank documents by cosine similarity to a query vector — the core retrieval mechanism inside every vector database.

[ ]

import numpy as np

def cosine_sim(a, b):
    return np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b))

store = {
    "cat":  np.array([0.90, 0.10, 0.20]),
    "dog":  np.array([0.80, 0.20, 0.30]),
    "car":  np.array([0.10, 0.90, 0.10]),
}
query = np.array([0.85, 0.15, 0.25])  # "feline pet?"

results = sorted(store.items(),
    key=lambda kv: cosine_sim(query, kv[1]),
    reverse=True)

for rank, (doc, vec) in enumerate(results, 1):
    print(f"{rank}. {doc}: {cosine_sim(query,vec):.3f}")

RankDocumentCosine similarityScore

Vector Databases — Quiz

1 Question

Question 1 of 1

Which data structure does a vector database use to enable fast approximate nearest-neighbor search?

Hash table HNSW graph B-tree index Inverted index

localhost:8501 — AI Chatbot Platform

Basic AI Chatgpt-4o

Hello! I am a basic AI assistant powered by GPT-4o. Ask me anything about data science or AI.

Search-Enabled Agentgpt-4o + Tavily

I can search the web for real-time information. What would you like to research?

Chat with your DataRAG Pipeline

agentic_ai_report_2025.pdf Indexed · 42 chunks

Document indexed and ready. I will ground every answer in its content. What would you like to know?

MCP AgentModel Context Protocol

calendarweb_searchfile_systememail

MCP tools connected: calendar, web_search, file_system, email. I can coordinate across all of them for complex tasks.

Why DSD Academy

Built for the real world,
not the classroom

Most online courses teach you to follow tutorials. We teach you to build systems that work in production.