--- title: Decoding the new AI lingo: Loops, harnesses, squads,... description: A glossary-style breakdown of buzzwords circulating in AI-assisted development, based on a GitHub Podcast episode. Covers loop engineering (repeatable agent... canonical: https://daily.dev/posts/decoding-the-new-ai-lingo-loops-harnesses-squads-hill-climbing-oh-my--uj0ibsge9 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Decoding the new AI lingo: Loops, harnesses, squads, hill climbing… oh my! | daily.dev og:description: A glossary-style breakdown of buzzwords circulating in AI-assisted development, based on a GitHub Podcast episode. Covers loop engineering (repeatable agent... og:url: https://daily.dev/posts/decoding-the-new-ai-lingo-loops-harnesses-squads-hill-climbing-oh-my--uj0ibsge9 og:image: https://api.daily.dev/og/posts/uj0iBSGe9.png og:image:alt: Decoding the new AI lingo: Loops, harnesses, squads, hill climbing… oh my! og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Decoding the new AI lingo: Loops, harnesses, squads, hill climbing… oh my! **[GitHub Blog](https://daily.dev/sources/ghblog)** · [@cassidoo](https://daily.dev/cassidoo) · 6 min read · 0 upvotes · 0 comments ## Summary A glossary-style breakdown of buzzwords circulating in AI-assisted development, based on a GitHub Podcast episode. Covers loop engineering (repeatable agent workflows), Ralph loops (brute-force iterative agent tasking), squads and fleets (multi-agent role specialization and parallelization), harnesses (the tooling, permissions, and orchestration surrounding a model, exemplified by GitHub Copilot), hill climbing (iteratively improving agents via evals and feedback), forward deployed engineers (customer-facing AI implementation roles), and the distinction between closed models, open weight models, and open source models based on what's shared with developers. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### what is the difference between a harness and a model when talking about AI coding agents A harness is everything surrounding an AI model that makes it usable in a workflow, including tools, permissions, memory, context, and orchestration, while the model itself only generates outputs. GitHub Copilot is an example of a harness because it connects a model to codebases, editors, pull requests, and terminals. Harness engineering refers to designing and improving that surrounding system. _daily.dev surfaces explainers like this for developers sorting out harnesses from the models they wrap._ ### what is a Ralph loop in AI agent workflows A Ralph loop is a pattern where you give an agent a detailed task, often from a product requirements document or spec, and let it keep working until the job is complete, cycling through plan-act-check steps. It's useful for breaking down large tasks but can be expensive and inefficient since every iteration burns more tokens, context, and compute. Loop engineering aims to structure this pattern more reliably. _Developers comparing agent-loop patterns can track how these terms evolve on daily.dev._ ### what is the difference between open weight models and open source models Open weight models release the trained model weights so developers can download and run them, often locally or on their own infrastructure, but the dataset and training method may not be fully disclosed. Open source models go further, making the model, code, data, and training process all available for inspection, reuse, and modification. Closed models, by contrast, are only accessible via an API or hosted product with no access to weights or training data. _Teams weighing model openness against control can keep up with these distinctions on daily.dev._ --- Tags: [#open-source](https://daily.dev/tags/open-source), [#github](https://daily.dev/tags/github), [#ai-agents](https://daily.dev/tags/ai-agents), [#prompt-engineering](https://daily.dev/tags/prompt-engineering) [View this post on daily.dev](https://daily.dev/posts/decoding-the-new-ai-lingo-loops-harnesses-squads-hill-climbing-oh-my--uj0ibsge9) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Decoding the new AI lingo: Loops, harnesses, squads, hill climbing… oh my!","url":"https://daily.dev/posts/decoding-the-new-ai-lingo-loops-harnesses-squads-hill-climbing-oh-my--uj0ibsge9","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/decoding-the-new-ai-lingo-loops-harnesses-squads-hill-climbing-oh-my--uj0ibsge9"},"datePublished":"2026-09-02T21:04:24.643Z","dateModified":"2026-09-03T18:00:15.108Z","description":"A glossary-style breakdown of buzzwords circulating in AI-assisted development, based on a GitHub Podcast episode. Covers loop engineering (repeatable agent...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/4d08d94399bcdf9cb3aa1f5002f34172?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/4d08d94399bcdf9cb3aa1f5002f34172?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"GitHub Blog","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Person","name":"Cassidy Williams","url":"https://daily.dev/cassidoo","image":"https://avatars.githubusercontent.com/u/1454517?v=4","description":"making memes and dreams and software","interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"EndorseAction"},"userInteractionCount":2180}},"commentCount":0,"discussionUrl":"https://daily.dev/posts/decoding-the-new-ai-lingo-loops-harnesses-squads-hill-climbing-oh-my--uj0ibsge9","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"open-source,github,ai-agents,prompt-engineering","timeRequired":"PT6M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"GitHub Blog","item":"https://daily.dev/sources/ghblog"},{"@type":"ListItem","position":3,"name":"Decoding the new AI lingo: Loops, harnesses, squads, hill climbing… oh my!"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/decoding-the-new-ai-lingo-loops-harnesses-squads-hill-climbing-oh-my--uj0ibsge9#faq","mainEntity":[{"@type":"Question","name":"what is the difference between a harness and a model when talking about AI coding agents","acceptedAnswer":{"@type":"Answer","text":"A harness is everything surrounding an AI model that makes it usable in a workflow, including tools, permissions, memory, context, and orchestration, while the model itself only generates outputs. GitHub Copilot is an example of a harness because it connects a model to codebases, editors, pull requests, and terminals. Harness engineering refers to designing and improving that surrounding system. daily.dev surfaces explainers like this for developers sorting out harnesses from the models they wrap."}},{"@type":"Question","name":"what is a Ralph loop in AI agent workflows","acceptedAnswer":{"@type":"Answer","text":"A Ralph loop is a pattern where you give an agent a detailed task, often from a product requirements document or spec, and let it keep working until the job is complete, cycling through plan-act-check steps. It's useful for breaking down large tasks but can be expensive and inefficient since every iteration burns more tokens, context, and compute. Loop engineering aims to structure this pattern more reliably. Developers comparing agent-loop patterns can track how these terms evolve on daily.dev."}},{"@type":"Question","name":"what is the difference between open weight models and open source models","acceptedAnswer":{"@type":"Answer","text":"Open weight models release the trained model weights so developers can download and run them, often locally or on their own infrastructure, but the dataset and training method may not be fully disclosed. Open source models go further, making the model, code, data, and training process all available for inspection, reuse, and modification. Closed models, by contrast, are only accessible via an API or hosted product with no access to weights or training data. Teams weighing model openness against control can keep up with these distinctions on daily.dev."}}]} ```