# Search and Discovery Published articles for Search and Discovery. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Mintlify Chat - ChatGPT Trained on Your Documentation DevFeed: [Mintlify Chat - ChatGPT Trained on Your Documentation]() Original publisher: [Read original article]() Author: Han Wang Published: 2023-07-19T00:00:00Z Content type: release Language: en Sources: [Mintlify Blog]() Topics: [Documentation](), [Artificial Intelligence](), [ChatGPT]() Tags: [ai](), [ai-trends](), [chatgpt](), [developer](), [documentation](), [hallucinations](), [information](), [quality](), [search-and-discovery]() ### AI overview Mintlify introduces Mintlify Chat, a GPT-4-powered search and answer system embedded in documentation. The article says it provides context-aware responses for developer onboarding, troubleshooting, and feature discovery, while addressing earlier concerns about GPT-3.5's inconsistency and hallucinations. ### Source excerpt Ever since Mintlify launched, we've been looking for ways to improve information discoverability. While GPT-3.5 had potential, it lacked consistency and quality, being prone to hallucinations and pulling out misleading or entirely false results from thin air. ## Search at Shopify--Range in Data and Engineering is the Future DevFeed: [Search at Shopify--Range in Data and Engineering is the Future]() Original publisher: [Read original article]() Author: Doug Turnbull Published: 2022-01-14T17:30:01Z Content type: opinion Language: en Sources: [Shopify Engineering](), [Shopify Engineering - Shopify Engineering]() Topics: [Shopify](), [Data Science](), [coding](), [machine learning research]() Tags: [data](), [data-science-and-engineering](), [engineering](), [search](), [search-and-discovery](), [shopify]() ### AI overview Shopify's search team treats range across data science and engineering as a core working principle. The article argues that combining both perspectives helps teams understand trade-offs, avoid silos, make better decisions, and deliver machine learning models to production. ### Source excerpt At Shopify, we draw very few lines between "data" and "engineering" work. Instead we have "search" work. ## Using Rich Image and Text Data to Categorize Products at Scale DevFeed: [Using Rich Image and Text Data to Categorize Products at Scale]() Original publisher: [Read original article]() Author: Kshetrajna Raghavan Published: 2021-09-08T14:30:00Z Content type: article Language: en Sources: [Shopify Engineering](), [Shopify Engineering - Shopify Engineering]() Topics: [Model Development](), [Machine Learning & Artificial Intelligence](), [data]() Tags: [data](), [image](), [machine-learning](), [model](), [precision](), [predictions](), [scale](), [search-and-discovery]() ### AI overview Shopify describes how it modernized its product categorization model using rich image and text data. The new model increased leaf precision by 8% while doubling coverage, while addressing scale, language, latency, and access-pattern requirements. ### Source excerpt We reevaluated our existing product categorization model to ensure we're understanding what our merchants are selling, to build the best products that help power their sales. ## Irrational Fun: Find Yourself at Berlin Buzzwords DevFeed: [Irrational Fun: Find Yourself at Berlin Buzzwords]() Original publisher: [Read original article]() Published: 2014-04-27T00:00:00Z Content type: article Language: en Sources: [SoundCloud Backstage Blog]() Topics: [Code](), [Algorithm](), [datasets](), [Code quality](), [Image]() Tags: [code](), [code-quality](), [contests](), [data](), [image](), [mapping](), [performance](), [search-and-discovery](), [source]() ### AI overview SoundCloud announces a Berlin Buzzwords contest in which participants must write an algorithm to find sequences of digits of π that most closely match a supplied grayscale logo image. Entries are evaluated on code quality, runtime performance, and visual similarity using a dataset containing the first billion digits of π. ### Source excerpt We were counting down the days until Berlin Buzzwords on May 25, when we realised that it would be great if you came too! With that in mind... ## Architecture behind our new Search and Explore experience DevFeed: [Architecture behind our new Search and Explore experience]() Original publisher: [Read original article]() Published: 2012-12-04T00:00:00Z Content type: article Language: en Sources: [SoundCloud Backstage Blog]() Topics: [App](), [real-time](), [User Experience](), [Network](), [Replication](), [Database]() Tags: [apache](), [app](), [architecture](), [consumer](), [data](), [database](), [experiment](), [explore](), [feature](), [features](), [infrastructure](), [models](), [network](), [new-feature](), [recovery](), [replication](), [scale](), [search](), [search-and-discovery](), [speed](), [user-experience]() ### AI overview SoundCloud describes the architecture behind its new Search and Explore experience. The article explains that the earlier Apache Solr search implementation became difficult to keep real-time, re-index, recover, scale, and modify as the product and data grew. The supplied text introduces an overhaul intended to improve relevance, scalability, experimentation, and feature delivery, but does not include the details of the new architecture. ### Source excerpt Search is front-and-center in the new SoundCloud, key to the consumer experience. We've made the search box one of the first things you see...