WeAreDevelopers Logo WeAreDevelopers
Sign in
For You LIVE
WeAreDevelopers LIVE • Feb 23, 2024

Harry Potter and the Elastic Semantic Search

Iulia Feroli

You don't need complex generative AI to build magical search experiences. Learn to construct a highly relevant, hybrid semantic search engine using Python, Elasticsearch, and Hugging Face.

Pause
Mute Enter Fullscreen
#1 about 4 min

Generating early AI text with natural language processing

Historical AI advancements trace the progression of rudimentary generative text processing capabilities.

#2 about 3 min

Understanding vector embeddings and text representation

Transforming unstructured phrases into multidimensional numeric arrays allows systems to calculate linguistic properties.

#3 about 2 min

Transforming text into mathematical features with algorithms

Quantifying language occurrences via algorithms produces vectors that encode dataset properties comprehensively.

#4 about 2 min

Mapping data relations through vector embedding spaces

Analyzing multidimensional proximity reveals contextual relationships between distinct words and concepts.

#5 about 2 min

Abstracting semantic rules and complex language intent

Programming models to map structural word relationships constructs a deeper understanding of linguistic intent.

#6 about 2 min

Enabling semantic search with automated query vectorization

Vectorizing incoming questions queries isolates contextually matching documents stored within the same embedding space.

#7 about 2 min

Customizing neural network inference through domain datasets

Feeding specific operational data into pre-trained transformer architecture refines relevant output accuracy natively.

#8 about 2 min

Distinguishing classic machine learning from generative AI

Separating deep learning layers from buzzword generalities clarifies when to deploy modern generative concepts.

#9 about 2 min

Utilizing Elasticsearch infrastructure as a vector database

Storing high-dimensional vectors natively equips platforms to handle advanced semantic queries efficiently.

#10 about 3 min

Integrating Hugging Face models for data inference

Deploying pre-trained open source models allows databases to perform dynamic classification tasks on user input easily.

#11 about 4 min

Demonstrating standard limitations of exact match searching

Examining strict syntax requirements highlights how rudimentary querying filters miss contextual variations.

#12 about 4 min

Executing database queries programmatically with Python clients

Transmitting nested operational search code through software endpoints builds resilient querying mechanisms reliably.

#13 about 4 min

Implementing sentiment analysis pipelines on ingested text

Adding model layers natively classifies indexable text inputs by underlying sentiment dynamically.

#14 about 4 min

Running nearest neighbor searches on vectorized data

Computing coordinate distances maps input queries against dense text indexes for contextual retrieval.

#15 about 4 min

Combining structured filters with semantic hybrid searches

Coupling strict string filters with semantic processing builds a highly refined response system.

#16 about 3 min

Reviewing system architecture for custom semantic search

Synthesizing disparate processing modules with established infrastructure anchors scalable enterprise retrieval ecosystems.

#17 about 3 min

Addressing relevance issues and natural language complexity

Validating search behaviors against query formulas dictates resulting outcome quality during natural conversation.

#18 about 2 min

Handling missing terminology definitions during hybrid execution

Overlapping semantic vectors with explicit string logic limits vocabulary disruption intuitively.

#19 about 2 min

Automating data pipeline updates for production environments

Configuring continual ingestion workflows prevents indexing disruptions across dynamic document infrastructures.

#20 about 2 min

Importing third-party language models into established pipelines

Linking secure modeling capabilities directly preserves search architectures while enriching output variance.

#21 about 2 min

Identifying appropriate use cases when implementing generative AI

Prioritizing objective software problems logically mitigates expensive computational waste associated with hyped solutions.

#22 about 3 min

Debugging outputs and analyzing text chunking strategies

Altering segmenting constraints iteratively identifies how formatting rules alter final algorithmic resolutions.

#23 about 2 min

Improving model result rankings alongside user feedback

Utilizing integrated telemetry structures scales search effectiveness precisely through directed human adjustment.

#24 about 2 min

Practical applications for targeted sentiment analysis queries

Isolating subjective emotion values empowers customer support routing to categorize urgent concerns organically.

#25 about 3 min

Evaluating domain-specific models against industry benchmarks

Testing distinct processing formats simultaneously reveals which models best digest localized business terminology.

Matching moments

1:23 min

Powering website search queries with generative language algorithms

Kapil Gupta Kapil Gupta · WWC 2023

2:38 min

Enabling contextual responses with retrieval-augmented generation and vector databases

David Leconte David Leconte +1 · WWC 2024

3:00 min

Introduction to generative AI and vector search

Mary Grygleski Mary Grygleski · LIVE

1:03 min

Revolutionizing database search queries with language models

Calvin Seward Calvin Seward · Europe 2026 Virtual

3:08 min

Scaling semantic search with Astra DB and Apache Cassandra

David Leconte David Leconte +1 · WWC 2024

3:46 min

Substituting traditional classification models with search-based AI architecture

Erik Bamberg · LIVE

More from WeAreDevelopers LIVE

Aliens in Quarantine, Python Deprecates Loops, and AI Buys Old Books - Zbyszek Tenerowicz thumbnail
5:43
5 days ago

Aliens in Quarantine, Python Deprecates Loops, and AI Buys Old Books - Zbyszek Tenerowicz

Chris Heilmann, Daniel Cranney

WeAreDevelopers LIVE - Node and Package Security thumbnail
1:08:30
5 days ago

WeAreDevelopers LIVE - Node and Package Security

Chris Heilmann, Daniel Cranney, Zbyszek Tenerowicz

WeAreDevelopers LIVE - Streaming HTML thumbnail
1:10:26
12 days ago

WeAreDevelopers LIVE - Streaming HTML

Chris Heilmann, Daniel Cranney, Jad Joubran

WeAreDevelopers LIVE - CODE100 at WeAreDevelopers Europe thumbnail
1:02:34
19 days ago

WeAreDevelopers LIVE - CODE100 at WeAreDevelopers Europe

Chris Heilmann, Daniel Cranney, Justin Halsall, Werner Jeschko, Nicole Jeschko, Mirco Tomić

WeAreDevelopers LIVE - Web Performance in an AI World thumbnail
1:07:59
26 days ago

WeAreDevelopers LIVE - Web Performance in an AI World

Chris Heilmann, Daniel Cranney

FIFA Hack, Midjourney Does Ultrasounds, and Trampoline Traffic Lights - Nicole Jeschko and Werner Jeschko thumbnail
5:55
about 2 months ago

FIFA Hack, Midjourney Does Ultrasounds, and Trampoline Traffic Lights - Nicole Jeschko and Werner Jeschko

Chris Heilmann, Daniel Cranney

Upcoming sessions on this topic

Open session

World Congress 2026 North America

Building Stuff with GenAI - The Open Minded Workshop beyond OpenAI

Andreas Erben

CTO for Applied AI and Metaverse at daenet

Andreas Erben
Open session

World Congress 2026 North America

Beyond SQL Generation: How to Teach Agents What Your Database Actually Means

Kris Jenkins

Lead Developer Advocate at Snowflake

Kris Jenkins
Open session

World Congress 2026 North America

Building Pragmatic AI: 10 AI Features Your Users Actually Want

Jonathan "J." Tower

.NET Foundation Board | 12x Microsoft MVP | Founder & Consultant

Jonathan "J." Tower
Open session

World Congress 2026 North America

No Single Model to Rule Them All: Building Resilient AI Agents Across Open & Closed LLMs

Emmanuel Acheampong

Senior Manager Developer Relations at Crusoe AI

Emmanuel Acheampong
Open session

World Congress 2026 North America

Making Science Larger, not just Faster

Yuval Dvir

Commercial Executive, SandboxAQ

Yuval Dvir
Open session

World Congress 2026 North America

You Can’t Re-Run Sunlight: Designing ML Data Architectures for Physical AI

An Phan

Senior Data Infrastructure Engineer @ Hippo Harvest

An Phan

Related videos

Hybrid AI: Next Generation Natural Language Processing thumbnail
21:01
WWC 2022 • about 4 years ago

Hybrid AI: Next Generation Natural Language Processing

Jan Schweiger

Enter the Brave New World of GenAI with Vector Search thumbnail
1:00:56
LIVE • over 2 years ago

Enter the Brave New World of GenAI with Vector Search

Mary Grygleski

Semantic AI: Why Embeddings Might Matter More Than LLMs thumbnail
26:05
WWC 2025 • 12 months ago

Semantic AI: Why Embeddings Might Matter More Than LLMs

Christian Weyer

How to Decipher User Uncertainty with GenAI and Vector Search thumbnail
24:47
WWC 2025 • 12 months ago

How to Decipher User Uncertainty with GenAI and Vector Search

Ben Greenberg

Develop AI-powered Applications with OpenAI Embeddings and Azure Search thumbnail
57:52
LIVE • over 2 years ago

Develop AI-powered Applications with OpenAI Embeddings and Azure Search

Rainer Stropek

Best practices: Building Enterprise Applications that leverage GenAI thumbnail
24:24
WWC 2025 • 12 months ago

Best practices: Building Enterprise Applications that leverage GenAI

Damir

Related articles

From learning to earning

Jobs that call for the skills explored in this talk.