Ever wondered what happens when data and AI sit down for a chat? "Data Meets AI" bridges the gap between the worlds of data engineering and artificial intelligence. In short, bite-sized episodes, learn how these two fields can learn from each other.
Data Meets AI
Claim This Podcastby Jorrit Sandbrink, Nicolay Christopher Gerold
2 episodes
Updated Weekly
Accepts GuestsHas SponsorsLocation 🇩🇰
Podcast Overview
Ever wondered what happens when data and AI sit down for a chat? "Data Meets AI" bridges the gap between the worlds of data engineering and artificial intelligence. In short, bite-sized episodes, learn how these two fields can learn from each other.
Language
🇺🇲
Publishing Since
6/10/2024
1 verified contact email on file for Data Meets AI
Pitch yourself as a guest, propose sponsorships, or reach out directly to the host.
Recent Episodes
June 10, 2024
What is RAG? AI for data engineers.
<p>RAG is a way to feed data into a language model (LLM) to improve its results. It involves retrieving relevant information based on a user query and using that information to fill out a template for the LLM.</p> <p>RAG can be performed through semantic or embedding search, as well as keyword-based or full-text search. The basic architecture of Rack includes a retrieval component (vector database) and a generation component (LLM), with the prompt serving as the template.</p> <p>It works well when the model lacks relevant information or when proprietary or resource languages are involved. It doesn't work well with new terms or irrelevant data.</p> <p><strong>Takeaways</strong></p> <p>A feature store is an interface between models and data, simplifying the process of working with old features.</p> <p>Feature tables are entity-based and contain one record per entity instance, with each feature as a column.</p> <p>The feature store solves problems like training-serving skew, data leakage, feature redundancy, and lack of lineage.</p> <p>It is useful for smaller teams and larger organizations that want to simplify their machine learning workflows.</p> <p><strong>Jorrit Sandbrink:</strong></p> <ul> <li><a href="https://www.linkedin.com/in/jorritsandbrink/">LinkedIn</a></li> </ul> <p><strong>Nicolay Gerold:</strong></p> <ul> <li><a href="https://www.linkedin.com/in/nicolay-gerold/">LinkedIn</a></li> <li><a href="https://twitter.com/nicolaygerold">X (Twitter)</a></li> </ul> <p><strong>00:00</strong> Introduction to Rack</p> <p><strong>01:34</strong> Adding Additional Information to Language Models</p> <p><strong>03:51</strong> When to Use Rack</p> <p><strong>06:35</strong> Limitations of Rack</p> <p><strong>08:14</strong> Tools for Building Rack</p> <p>Keywords: RAG, data engineering, AI, language model, LLM, retrieval, generation, semantic search, embedding search, keyword-based search, full-text search, architecture, prompt, knowledge base, tools</p> --- Send in a voice message: https://podcasters.spotify.com/pod/show/nicolay-gerold/message
June 10, 2024
What is a feature store? A data engineering perspective.
<p>A feature store is an interface between models and data, serving as an abstraction layer built on top of existing infrastructure. It is not a new database type but rather a way to simplify working with old features. </p> <p>The store store consists of feature tables, which are entity-based and contain one record per entity instance. The main interactions with a feature table are feature engineering, model training, and inference. It solves problems like training-serving skew, data leakage, feature redundancy, and lack of lineage. </p> <p><strong>Takeaways</strong></p> <p>A feature store is an interface between models and data, simplifying the process of working with old features.</p> <p>Feature tables are entity-based and contain one record per entity instance, with each feature as a column.</p> <p>The feature store solves problems like training-serving skew, data leakage, feature redundancy, and lack of lineage.</p> <p>It is useful for smaller teams and larger organizations that want to simplify their machine learning workflows.</p> <p><strong>Jorrit Sandbrink:</strong></p> <ul> <li><a href="https://www.linkedin.com/in/jorritsandbrink/" target="_blank" rel="noopener noreferer"><strong>LinkedIn</strong></a></li> </ul> <p><strong>Nicolay Gerold:</strong></p> <ul> <li><a href="https://www.linkedin.com/in/nicolay-gerold/">LinkedIn</a></li> <li><a href="https://twitter.com/nicolaygerold">X (Twitter)</a></li> </ul> <p>00:00 Understanding the Concept of a Feature Store</p> <p>05:25 The Structure of a Feature Store: Feature Tables and Interactions</p> <p>09:04Problems Solved by a Feature Store</p> <p>11:46 Target Users of Feature Stores</p> <p>16:11 Offline vs. Online Feature Stores</p> <p>Keywords: feature store, abstraction layer, feature table, feature engineering, model training, inference, training-serving skew, data leakage, feature redundancy, lack of lineage</p> --- Send in a voice message: https://podcasters.spotify.com/pod/show/nicolay-gerold/message
2 total episodes available
Deep-dive analytics for Data Meets AI
Frequently asked questions
Have a different question and can't find the answer you're looking for? Reach out to our support team by sending us an email and we'll get back to you as soon as we can.
- What is Data Meets AI?
- How often does this podcast release new episodes?
This podcast updates weekly.
- Where can I listen to this podcast?
This podcast is available on 6 platforms including Apple Podcasts, Spotify, and more. You can also use the RSS feed directly.
- Does this podcast accept guests?
Information about guest appearances is not available.
Legal Disclaimer
Pod Engine is not affiliated with, endorsed by, or officially connected with any of the podcasts displayed on this platform. We operate independently as a podcast discovery and analytics service.
All podcast artwork, thumbnails, and content displayed on this page are the property of their respective owners and are protected by applicable copyright laws. This includes, but is not limited to, podcast cover art, episode artwork, show descriptions, episode titles, transcripts, audio snippets, and any other content originating from the podcast creators or their licensors.
We display this content under fair use principles and/or implied license for the purpose of podcast discovery, information, and commentary. We make no claim of ownership over any podcast content, artwork, or related materials shown on this platform. All trademarks, service marks, and trade names are the property of their respective owners.
While we strive to ensure all content usage is properly authorized, if you are a rights holder and believe your content is being used inappropriately or without proper authorization, please contact us immediately at hey@podengine.ai for prompt review and appropriate action, which may include content removal or proper attribution.
By accessing and using this platform, you acknowledge and agree to respect all applicable copyright laws and intellectual property rights of content owners. Any unauthorized reproduction, distribution, or commercial use of the content displayed on this platform is strictly prohibited.