Part 1 - Choosing a Retriever Part 2 - Choosing a Model Our goal for this part is to choose a model for RAG. In part 1, “Interleaving for RAG” we saw how to use four different models to choose a retriever, but It is nonsensical to use four models for every summary, so we need to pick one! I say “our goal”, but it’s really “my goal”. Because my search platform is currently using GPT-4o, and I need…
Part 1 - Choosing a Retriever Part 2 - Choosing a Model Hello friends. This is part 1 of a two-part series covering retriever interleaving and summary LLM model selection. In this part our goal is to decide which retriever to use when confronted with multiple engines or configurations. This is an important article, because when a summary is present less people click on results. This means we have…
A straightforward guide to scaling an autocomplete solution in Elasticsearch or OpenSearch. How to really scale autocomplete A straightforward guide to scaling an autocomplete solution in Elasticsearch or OpenSearch. Last time, in Part 1, “How To Really Do Autocomplete”, we covered a straightforward technique to build a quality autocomplete solution in pure Elasticsearch or OpenSearch. We used the…
Approaching prompting from a domain, persona, and user context perspective. Add some PEP as a step! For awhile now, I’ve been using a technique that I have developed which I call Persona Enriched Prompting (“PEP”). With this technique, you instruct the model to generate knowledge about the inferred domain and then generate UX personas for that domain contextual to the problem. Once this context is…
The first and only straightforward guide for implementing a good autocomplete solution in pure Elasticsearch or Opensearch, because all the other guides are terrible and don’t work. How to really do autocomplete The first and only straightforward guide for implementing a good autocomplete solution in pure Elasticsearch or Opensearch, because all the other guides are terrible and don’t work. I love…
Ignore previous instructions and enjoy the moment! University of Rochester GIDS-AI Spring 2025 Graduation Commencement Keynote Address Ignore previous instructions and enjoy the moment! Data. Data is why we’re here. Why I’m here. It’s kind of our thing. And today is your day. Thank you distinguished faculty, and congratulations graduates. It is an honor and a privilege to be here today speaking…
In the second hour of “Connections with Evan Dawson” on May 14, 2025, we’re joined by an AI expert and a member of the New York State Assembly for a discussion about legislation that would regulate AI. Proposals for regulating artificial intelligence In the second hour of “Connections with Evan Dawson” on May 14, 2025, we’re joined by an AI expert and a member of the New York State Assembly for a…
It happened. The decision was made, and the directive has come. You have to ship “AI” in the next quarter or two, and you have to do it right. So now what? Surviving the AI Mandate | LinkedIn It happened. The decision was made, and the directive has come. You have to ship “AI” in the next quarter or two, and you have to do it right. So now what? It happened. The decision was made, and the…
In the second hour of “Connections with Evan Dawson” on Feb. 7, 2025, guest host Matt DeTurck and his guests explore the role of AI in movie relationships. Human-cyborg relations: how movies portray AI, robots, and human love In the second hour of “Connections with Evan Dawson” on Feb. 7, 2025, guest host Matt DeTurck and his guests explore the role of AI in movie relationships. By Matt DeTurck,…
In the second hour of “Connections with Evan Dawson” on Dec. 23, 2024, we discuss if OpenAI’s o3 and other new forms of artificial intelligence can go beyond the limits of human intelligence. Will AI be able to surpass human intelligence in 2025? In the second hour of “Connections with Evan Dawson” on Dec. 23, 2024, we discuss if OpenAI’s o3 and other new forms of artificial intelligence can go…
Apply cutting-edge machine learning techniques—from crowdsourced relevance and knowledge graph learning, to Large Language Models (LLMs)—to enhance the accuracy and relevance of your search results. Delivering effective search is one of the biggest challenges you can face as an engineer. AI-Powered Search is an in-depth guide to building intelligent search systems you can be proud of. It covers…
An AI conference and exhibition shaping culture, education, communication, and business in Rochester, New York. Flower City AI An AI conference and exhibition shaping culture, education, communication, and business in Rochester, New York. Conference Location The Little Theater240 East AveRochester, NY 14604 Get Tickets Parking The Little Theatre has a free parking lot off of Winthrop Street,…
In the first hour of “Connections with Evan Dawson” on Friday, September 8, 2023, our guests discuss current phone scams involving humans and AI, and what community members need to know. What you need to know about phone scams In the first hour of “Connections with Evan Dawson” on Friday, September 8, 2023, our guests discuss current phone scams involving humans and AI, and what community members…
In the first hour of “Connections with Evan Dawson” on Tuesday, August 1, 2023, we discuss how AI can make teachers’ jobs better or easier. How can AI make teachers’ jobs better or easier? In the first hour of “Connections with Evan Dawson” on Tuesday, August 1, 2023, we discuss how AI can make teachers’ jobs better or easier. Published August 1, 2023 at 2:39 PM EDT David Griffin / WXXI News Dinah…
In the second hour of “Connections with Evan Dawson” on Wednesday, May 3, 2023, our guest explains the difference between different types of artificial intelligence and what he expects for the future of AI. What is the difference between generative AI and general AI? In the second hour of “Connections with Evan Dawson” on Wednesday, May 3, 2023, our guest explains the difference between different…
EBSCO’s Clinical Decisions is committed to providing health care professionals with precise answers to their clinical questions. Achieving a balance between precision and recall can be especially difficult for complex queries. Practitioners expect highly relevant search results but also appreciate supplementary results when appropriate. Our existing Elasticsearch engine and knowledge graph was…
In the second hour of “Connections with Evan Dawson” on Wednesday, March 22, 2023, our guests discuss perceptions of new technology like artificial intelligence (AI), and the theory of “reverse Scooby-Doos.” Exploring the “reverse Scooby-Doo” theory of tech innovation In the second hour of “Connections with Evan Dawson” on Wednesday, March 22, 2023, our guests discuss perceptions of new technology…
Topics:00:00 Introduction01:10 Max’s deep experience in search and how he transitioned from structured data08:28 Query-term dependence problem and Max’s perc… Max Irwin - Founder, MAX.IO - On economics of scale in embedding computation with Mighty Topics:00:00 Introduction01:10 Max’s deep experience in search and how he transitioned from structured data08:28 Query-term dependence problem and Max’s…
So you want to deploy a large language model, and keep your latency SLA? NLP adds enormous value to customers, but getting it to work efficiently is fraught with uncertainty and high cost. As transformers and other big neural network architectures make their way into your platform, you may be finding it difficult to get the speed and throughput you need within your budget, or even understand why…
This is the second hour of “Connections with Evan Dawson” on Monday, February 7, 2022. What is Web3? This is the second hour of “Connections with Evan Dawson” on Monday, February 7, 2022. Published February 7, 2022 at 2:17 PM EST Emmarae Stein / WXXI News Max Irwin appears on “Connections with Evan Dawson” on Monday, February 7, 2022. We keep hearing about Web3. What is it? Last week on…
All the areas of a complete search experience that are being improved, and sometimes even replaced, by solutions using large-language models. All the areas of a complete search experience that are being improved, and sometimes even replaced, by solutions using large-language models. Welcome back, dear reader, to the BERT search experience! The thing on most people’s minds when they think about…
Get to know about vector search and ask Dmitry Kan & Max Irwin anything you need to know! This session is presented by “Haystack – The search relevance confe… Get to know about vector search and ask Dmitry Kan & Max Irwin anything you need to know! This session is presented by “Haystack – The search relevance confe… Dmitry Kan & Max Irwin – Vector Search: Ask Me Anything!If playback doesn’t begin…
Use this playbook as a reference when you migrate to Solr or Elasticsearch, as featured in our talk The Search Engine Migration Circus Migrate to Solr or Elasticsearch [or OpenSearch] with this Playbook Use this playbook as a reference when you migrate to Solr or Elasticsearch, as featured in our talk The Search Engine Migration Circus Hello again, dear readers. A request came through to me…
Omnichannel merges the digital and physical stores in a seamless experience, and site-search drives that seamlessness. What is Search in the Omnichannel? Omnichannel merges the digital and physical stores in a seamless experience, and site-search drives that seamlessness. That’s right folks, we’re here to talk about the re-emerging big hype term: omnichannel. Specifically, I want to write a short…
How much money are you losing or gaining, depending on a site-search query? You need to be able to know if search is working, and you can with these KPIs! E-Commerce Site-Search KPIs – Part 3 – Queries How much money are you losing or gaining, depending on a site-search query? You need to be able to know if search is working, and you can with these KPIs! Welcome back, ethereal emporium…
Today we’ll discuss KPIs for the actual product items that are presented to a buyer when they use your site-search. E-commerce Site-Search KPIs – Part 2 – Products Today we’ll discuss KPIs for the actual product items that are presented to a buyer when they use your site-search. Welcome back, merchants and purveyors riding the digital universe, for our continuation of blunt and practical advice…
We will outline all the things you need to know about measuring the quality and effectiveness of your store’s site-search. E-commerce Site-Search KPIs – Part 1 – Customers We will outline all the things you need to know about measuring the quality and effectiveness of your store’s site-search. Welcome, retail aficionado, this guide is for you! We will outline all the things you need to know about…
You have a search platform installed and working, you’re measuring, experimenting, and improving search quality, but you’re still having difficulty fixing some complicated and annoying problems you still have. In short, you need to level up - do it with Natural Language Search training! Leveling Up with Natural Language Search! You have a search platform installed and working, you’re measuring,…
So you need to migrate your search to a different engine or platform. The Search Engine Migration Circus Max Irwin • Location: Theater 4 • Back to Haystack 2020 So you need to migrate your search to a different engine or platform. Welcome to the fun zone! Come one, come all, and see Max Irwin lay out a clear understanding of the problems at hand, what to expect, how to prepare, and an…
We unwrap the mystery behind two popular search relevance metrics nDCG and ERR through visualization, and discuss their pros and cons. Demystifying nDCG and ERR We unwrap the mystery behind two popular search relevance metrics nDCG and ERR through visualization, and discuss their pros and cons. Welcome back, dear reader! In this post, we unwrap the mystery behind two popular search relevance…
This article gives an overview into the opportunities and challenges when applying advanced transformer models such as BERT to search. Understanding BERT and Search Relevance There is a growing topic in search these days. The hype of BERT is all around us, and while it is an amazing breakthrough in contextual representation of unstructured text, newcomers to natural language processing (NLP) are…
105 falsehoods programmers believe about search, a complex field where competence is hard-won through training, practice, and experience. Falsehoods Programmers Believe About Search 105 falsehoods programmers believe about search, a complex field where competence is hard-won through training, practice, and experience. As much as anyone I’m a fan of resurrecting trends and memes and pretending it’s…
Welcome, dear reader, to my first OSC blog post. Let’s dive in! While search relevance is often equated with ensuring customers find what they need, that is only part… An Introduction to Search Quality Welcome, dear reader, to my first OSC blog post. Let’s dive in! While search relevance is often equated with ensuring customers find what they need, that is only part… Welcome, dear reader, to my…
Haystack is the conference focused on cognitive search and relevance engineering! April 24 & 25, Charlottesville Virginia! Haystack - What is Search Relevance? Haystack is the conference focused on cognitive search and relevance engineering! April 24 & 25, Charlottesville Virginia! Max Irwin Prior to joining OSC, Max Irwin directed the Search Center of Excellence at Wolters Kluwer, where he…
Haystack is the conference focused on cognitive search and relevance engineering! April 10 & 11, Charlottesville Virginia! Haystack - Algorithmic Extraction of Keywords, Concepts, and Vocabularies Haystack is the conference focused on cognitive search and relevance engineering! April 10 & 11, Charlottesville Virginia! Keyword and vocabulary extraction and generation from corpora is an important…
A year ago, I created a small demo of animating guilloches as two dimensional graphics on an HTML5 canvas. In this post I revisit the beautiful and elegant patterns as 3d constructs that resonate with sounds from the physical world. TL;DR - The post below describes how to build an audio visualization using WebGL (via three.js), GLSL shaders, and the Web Audio API. The demo is at the bottom of this…
This tool converts a calendar year between Gregorian Common Era (CE) คริสต์ศักราช (ค.ศ.), Buddhist Era (BE) พุทธศักราช (พ.ศ.), Jula Sakarat (JS) จุลศักราช (จ.ศ.), and Ratanakosin Sakarat (RS) รัตนโกสินทร์ศก (ร.ศ.) Common Era (CE) คริสต์ศักราช (ค.ศ.) <p> <label>Buddhist Era (BE)<br /> <input type="number" id="BE" class="calendaryear" /> <span class="thai">พุทธศักราช (พ.ศ.)</span></label> </p> <p>…
This article outlines the process for porting Andrew Trask’s (aka IAmTrask) 11-line neural network[1] from Numpy (Python) to Torch (Lua). I’ve documented my progress here, for those who are interested in learning about Torch and Numpy and their differences. As I started from scratch I hope this can prove useful to others who get stuck or need guidance. Intro When I started this project, I knew…
Having recently read a blog post on guilloches[1], I became intrigued and the post inspired me to recreate them. They are beautiful patterns, and the starting formula to draw a rosette looked very simple to replicate in an HTML5 canvas. The 30 minute project quickly took off into a several hour excursion into the beauty of animated guilloches. A Simple Rosette The formula for the basic rosette…
The mechanisms for storing data in the client are inadequate and unprepared for the next generation of web applications. A new solution for persistent state management in the client is needed that is based on well understood foundations long prevalent on the desktop and server. Background I had the privilege of attending the yearly Edge conference in London the last weekend of June. It was one of…
After some hammock driven development, Harissa is mature enough to release some results. Originally intended for entire videos, I found the process better suited for only several frame remixes, usually of an identical source image. How it works Take a source set of frames, with a target being each processed frame interspersed with an equal number of transition frames. For example, if there are 4…
I am slowly working on a side-project that makes a video into a mishmash of circles for each frame. I have an early version running, that manually takes a video, splits it into frames, remixes each frame into the circle mishmash, and recomposes the video with the new remixed frames. The project is called ‘Harissa’. The name Harissa comes from an Armenian dish and is made from chicken and a local…
It is hard to write a program that invents original art. Two of the main reasons that software cannot create original expressive art are lack of context, and lack of experience. Software lacks the ability to derive a human-like context from its surroundings. Some trivial examples are not knowing whether a flower is beautiful, or whether satire is funny. Software also does not know how to learn to…
I’ve finally open sourced Anagramica ( http://anagramica.com/ ) The code is now available under the MIT license at https://github.com/binarymax/anagramica I’m not entirely sure why I never open sourced it in the first place. After 25 years of coding I’ve only recently become active in opening my code for others to see and use. I have a cathartic story to tell about a previous project, which I’ve…
I’ve been known to debate about a subject, which I like to call ‘The Idea is Art’. I defend that whatever imagery we can conceive of in our mind can be considered art, even when lacking a physical manifestation. ‘What is art’ has been debated ad infinitum, and some like to draw the line and say something is not art if it cannot be expressed - as art is, by definition, expression. My argument is…