{"author":{"@type":"Person","name":"Alexander Bock","url":"https://abxk.ai/en/autor/alexander-bock/"},"categories":["AI Development"],"content":{"markdown":"Ask three people what \"AI development\" means, and you may hear three different answers. A programmer thinks of an assistant that suggests code in the editor. A product manager thinks of a chatbot that answers customer questions. A researcher thinks of training a model on huge amounts of data. All three are right, but they describe different work, with different skills, tools and costs.\n\nThis is part 1 of Level 0 in our AI Development learning path. You need no programming knowledge. You will learn the three ways to do AI development and which roles work on each. You will also see how this path is built, so you can decide where to start.\n\n## Three meanings of \"AI development\"\n\nThe confusion clears up with one question: where does the AI do its work? It can help you while you build software. It can be a part of the product you build. Or it can be the thing you build.\n\nAt the center of all three is a **model**: the trained core of an AI system. It has learned patterns from large amounts of data and turns an input, such as a question, into an output, such as an answer. Most of this path is about **large language models** (LLMs): models trained on text that can also write code.\n\n![Three numbered cards: AI helps you build, you build on a model, you change the model; an arrow below shows that each way works closer to the model](/images/learn/en/what-is-ai-development-three-ways.webp \"Three ways to do AI development: from AI as a tool to the model itself.\")\n\n1. **Way 1: using AI to build software.** An assistant helps you write code. The finished product may contain no AI at all.\n2. **Way 2: building with AI.** Your product uses a finished model while people work with it, for example to answer their questions.\n3. **Way 3: building the AI itself.** You train a model or adapt an existing one with your own data.\n\n{{\u003c callout type=\"tip\" title=\"The one-click test\" \u003e}}\nImagine a user clicks a button in your finished product. Does a model produce the answer at that moment? If yes, you are building with AI (Way 2 or 3). If no, AI was your tool while you built it (Way 1).\n{{\u003c /callout \u003e}}\n\n## Way 1: AI helps you write software\n\nAn **AI coding assistant** works inside your code editor or next to it. You will meet it in three main forms:\n\n- **Autocomplete** suggests the next lines while you type.\n- **Chat** explains code, finds errors and writes small functions on request.\n- **Agents** take on a whole task, change several files, run tests and show you the result.\n\nSome products go one step further. App builders such as Emergent create a web or mobile app from a description in plain language, according to the vendor. This style of work is often called *vibe coding*: you describe what you want, the AI writes the code, and you test and correct it.\n\nThe result is ordinary software. It runs without a model. But the responsibility stays with you: assistants can use outdated libraries, call functions that do not exist or introduce security flaws. Someone has to read, test and understand the code before it goes live. Part 5 of this level explains [what AI coding assistants do and where they fail](/en/ai-development/learn/what-ai-coding-assistants-do/).\n\n## Way 2: You build a product on top of a model\n\nThis is what most people mean when they talk about building AI apps, and it is the focus of this learning path. You do not train anything. You take a finished model and connect it to your own product.\n\nYour product talks to the model through an **API** (application programming interface): a fixed way for two programs to exchange data over the internet. Your app sends a request with instructions and the user's question. The provider's server sends back the model's answer. Most providers bill this usage per **token**. A token is a small piece of text, such as a short word or part of a word. Part 3 of this level explains [APIs, tokens and context windows](/en/ai-development/learn/apis-tokens-and-context-windows/).\n\n{{\u003c callout type=\"example\" \u003e}}\nAn online shop wants an assistant that answers questions about delivery and returns. The team picks a model and writes instructions: \"Answer only from our help pages, in a friendly tone.\" They connect the help pages as a source and add a chat window to the website. Before launch, they test it with real customer questions.\n{{\u003c /callout \u003e}}\n\nYou can build such a product with code or without it:\n\n- **With code**, a developer calls the model's API from their own software. The app runs on a server or on a hosting platform such as Netlify.\n- **Without code**, you use a **no-code platform**: a tool where you build by clicking, filling in forms and writing instructions. According to the vendors, MindStudio lets you build your own AI agents without code. Devs.ai lets teams build and run AI apps and agents on their own data.\n\n## Way 3: You train or adapt the model itself\n\nHere the model itself is your work, at two very different scales:\n\n- **Training from scratch** means creating a new model from huge amounts of data. Many specialized chips run for weeks or months to do it. The large models behind today's AI assistants come from a relatively small number of companies and research labs.\n- **Fine-tuning** means you continue training an existing model with your own examples. It then learns a format, a style or a narrow task. This needs far less data and computing power than training from scratch.\n\nTo fine-tune a model yourself, you need its weights. The **weights** are the numbers a model learned during training. A model with **open weights** can be downloaded. You can run and adapt it on your own hardware or on rented **GPUs**, for example from a GPU cloud such as Runpod. GPUs are graphics processors, the chips most AI models run on. Check the license first: open weights do not always allow every use. Models with closed weights stay with their provider, and some providers offer fine-tuning of selected models as a service. Our [LLM database](/en/llm/) lists {{\u003c stat \"models\" \u003e}} models, and {{\u003c stat \"models-open\" \u003e}} of them have open weights.\n\nStill, fine-tuning is rarely the first step. Many problems that seem to need it are solved faster with better instructions. Another option is to look up the right passages in your documents and give them to the model with each question. This method is called **retrieval-augmented generation** (RAG). For facts that change often, such as product details, it is easier to keep documents up to date than to fine-tune again. Our Level 3 guide [Fine-Tuning vs. RAG vs. Prompting](/en/ai-development/learn/fine-tuning-vs-rag-vs-prompting/) helps you decide.\n\n## Roles: who does what\n\nDifferent roles work on the three ways. Two questions place each of them: do you write code, and do you change the model?\n\n![Two-by-two grid: without code and with a finished model, the no-code builder; with code, the developer; without code and changing the model, the domain expert; with code, the ML engineer](/images/learn/en/what-is-ai-development-roles.webp \"Four roles, placed by two questions: do they write code, and do they change the model?\")\n\nThe table shows what each role builds and which skills come first:\n\n| Role | What they build | First skills to learn | Typical tools |\n|---|---|---|---|\n| No-code builder | Assistants, agents and simple apps | Describe a task precisely, write instructions, test with real examples | No-code platforms, app builders |\n| Developer (sometimes called AI engineer) | AI features inside software, connected through APIs | One programming language, HTTP and JSON, keeping API keys secret, testing | Code editor with an AI assistant, model APIs, hosting |\n| ML engineer (machine learning) | Trained and fine-tuned models, data pipelines, model serving | Python, statistics, data preparation, evaluation | GPU cloud, training frameworks |\n| Domain expert | Example data, test cases and quality rules | Describe good results, collect examples, judge answers | Spreadsheets, test sets |\n\nJob titles differ from company to company, and in small teams one person often has several roles. A marketing manager may build a no-code assistant one week and collect test questions for the developers the next. What matters is the skill a task needs, not the title. Within Way 2, you can build with no code, with low code (visual tools plus a little code) or with full code. Part 4 of this level compares [no-code, low-code and code](/en/ai-development/learn/no-code-low-code-or-code/).\n\n## What this learning path covers\n\nThe path has five levels with five parts each.\n\n![A staircase of five levels: Start here, Beginner, Intermediate, Advanced and Expert, each with its main topics; Level 0 is highlighted](/images/learn/en/what-is-ai-development-path.webp \"The five levels of the AI Development learning path. You are at Level 0.\")\n\nMost parts follow Way 2: products built on top of finished models. Way 1 comes up in the parts on coding assistants (Level 0) and on a safe AI-assisted coding workflow (Level 2). Way 3 comes up in Level 3, with fine-tuning and the GPU infrastructure it needs. Level 4 then shows how to run open-weight models on your own hardware.\n\nOur [tool comparison for AI Development](/en/ai-development/#tools) lists {{\u003c stat \"tools\" \"ai-development\" \u003e}} tools for building, hosting and automating. Five of them fit the three ways well. Emergent fits Way 1. MindStudio and Devs.ai fit Way 2 without code. Netlify hosts web apps, and Runpod rents out GPUs for Way 3.\n\n{{\u003c tools \"emergent mindstudio devs-ai netlify runpod\" \u003e}}\n\n## Find your starting point\n\nYour first project tells you where to start. Two questions point you to a way. Before Way 3, one more test can save you a lot of work.\n\n![Decision flow: if the finished product does not use AI itself, Way 1; if it does and the model does not have to learn anything new, Way 2; if the model must learn a new skill, Way 3, after testing prompts and RAG first](/images/learn/en/what-is-ai-development-decide.webp \"Two questions lead to your way. Test prompts and documents before you choose Way 3.\")\n\n| Your goal | Your way | Start with |\n|---|---|---|\n| Write and understand code faster | Way 1 | Level 0, part 5: AI coding assistants |\n| Build an internal assistant without code | Way 2, no code | Level 1, part 3: a simple AI app without code |\n| Add an AI feature to your own software | Way 2, with code | Level 1, part 1: your first LLM API call |\n| Teach a model a house style that prompts alone cannot hold | Way 3 | Level 3, part 4: fine-tuning, RAG or prompting |\n\n{{\u003c callout type=\"try\" \u003e}}\nTake one AI idea from your own work, for example \"answer supplier questions about our delivery terms\". Write down your answers:\n\n1. Will the finished product use a model while people work with it?\n2. Are good instructions and your documents enough, or must the model learn something new?\n3. Who could build it: someone without code, a developer or an ML engineer?\n4. Which way follows, and which part of this path will you read next?\n\nKeep the note. After Level 1, check whether your answers still hold.\n{{\u003c /callout \u003e}}","text":"Ask three people what \u0026ldquo;AI development\u0026rdquo; means, and you may hear three different answers. A programmer thinks of an assistant that suggests code in the editor. A product manager thinks of a chatbot that answers customer questions. A researcher thinks of training a model on huge amounts of data. All three are right, but they describe different work, with different skills, tools and costs.\nThis is part 1 of Level 0 in our AI Development learning path. You need no programming knowledge. You will learn the three ways to do AI development and which roles work on each. You will also see how this path is built, so you can decide where to start.\nThree meanings of \u0026ldquo;AI development\u0026rdquo; The confusion clears up with one question: where does the AI do its work? It can help you while you build software. It can be a part of the product you build. Or it can be the thing you build.\nAt the center of all three is a model: the trained core of an AI system. It has learned patterns from large amounts of data and turns an input, such as a question, into an output, such as an answer. Most of this path is about large language models (LLMs): models trained on text that can also write code.\nThree ways to do AI development: from AI as a tool to the model itself. Way 1: using AI to build software. An assistant helps you write code. The finished product may contain no AI at all. Way 2: building with AI. Your product uses a finished model while people work with it, for example to answer their questions. Way 3: building the AI itself. You train a model or adapt an existing one with your own data. ✦The one-click test\nImagine a user clicks a button in your finished product. Does a model produce the answer at that moment? If yes, you are building with AI (Way 2 or 3). If no, AI was your tool while you built it (Way 1).\nWay 1: AI helps you write software An AI coding assistant works inside your code editor or next to it. You will meet it in three main forms:\nAutocomplete suggests the next lines while you type. Chat explains code, finds errors and writes small functions on request. Agents take on a whole task, change several files, run tests and show you the result. Some products go one step further. App builders such as Emergent create a web or mobile app from a description in plain language, according to the vendor. This style of work is often called vibe coding: you describe what you want, the AI writes the code, and you test and correct it.\nThe result is ordinary software. It runs without a model. But the responsibility stays with you: assistants can use outdated libraries, call functions that do not exist or introduce security flaws. Someone has to read, test and understand the code before it goes live. Part 5 of this level explains what AI coding assistants do and where they fail.\nWay 2: You build a product on top of a model This is what most people mean when they talk about building AI apps, and it is the focus of this learning path. You do not train anything. You take a finished model and connect it to your own product.\nYour product talks to the model through an API (application programming interface): a fixed way for two programs to exchange data over the internet. Your app sends a request with instructions and the user\u0026rsquo;s question. The provider\u0026rsquo;s server sends back the model\u0026rsquo;s answer. Most providers bill this usage per token. A token is a small piece of text, such as a short word or part of a word. Part 3 of this level explains APIs, tokens and context windows.\n“Example\nAn online shop wants an assistant that answers questions about delivery and returns. The team picks a model and writes instructions: \u0026ldquo;Answer only from our help pages, in a friendly tone.\u0026rdquo; They connect the help pages as a source and add a chat window to the website. Before launch, they test it with real customer questions.\nYou can build such a product with code or without it:\nWith code, a developer calls the model\u0026rsquo;s API from their own software. The app runs on a server or on a hosting platform such as Netlify. Without code, you use a no-code platform: a tool where you build by clicking, filling in forms and writing instructions. According to the vendors, MindStudio lets you build your own AI agents without code. Devs.ai lets teams build and run AI apps and agents on their own data. Way 3: You train or adapt the model itself Here the model itself is your work, at two very different scales:\nTraining from scratch means creating a new model from huge amounts of data. Many specialized chips run for weeks or months to do it. The large models behind today\u0026rsquo;s AI assistants come from a relatively small number of companies and research labs. Fine-tuning means you continue training an existing model with your own examples. It then learns a format, a style or a narrow task. This needs far less data and computing power than training from scratch. To fine-tune a model yourself, you need its weights. The weights are the numbers a model learned during training. A model with open weights can be downloaded. You can run and adapt it on your own hardware or on rented GPUs, for example from a GPU cloud such as Runpod. GPUs are graphics processors, the chips most AI models run on. Check the license first: open weights do not always allow every use. Models with closed weights stay with their provider, and some providers offer fine-tuning of selected models as a service. Our LLM database lists 343 models, and 172 of them have open weights.\nStill, fine-tuning is rarely the first step. Many problems that seem to need it are solved faster with better instructions. Another option is to look up the right passages in your documents and give them to the model with each question. This method is called retrieval-augmented generation (RAG). For facts that change often, such as product details, it is easier to keep documents up to date than to fine-tune again. Our Level 3 guide Fine-Tuning vs. RAG vs. Prompting helps you decide.\nRoles: who does what Different roles work on the three ways. Two questions place each of them: do you write code, and do you change the model?\nFour roles, placed by two questions: do they write code, and do they change the model? The table shows what each role builds and which skills come first:\nWhat Is AI Development? Using AI vs. Building with AI Role What they build First skills to learn Typical tools No-code builder Assistants, agents and simple apps Describe a task precisely, write instructions, test with real examples No-code platforms, app builders Developer (sometimes called AI engineer) AI features inside software, connected through APIs One programming language, HTTP and JSON, keeping API keys secret, testing Code editor with an AI assistant, model APIs, hosting ML engineer (machine learning) Trained and fine-tuned models, data pipelines, model serving Python, statistics, data preparation, evaluation GPU cloud, training frameworks Domain expert Example data, test cases and quality rules Describe good results, collect examples, judge answers Spreadsheets, test sets Job titles differ from company to company, and in small teams one person often has several roles. A marketing manager may build a no-code assistant one week and collect test questions for the developers the next. What matters is the skill a task needs, not the title. Within Way 2, you can build with no code, with low code (visual tools plus a little code) or with full code. Part 4 of this level compares no-code, low-code and code.\nWhat this learning path covers The path has five levels with five parts each.\nThe five levels of the AI Development learning path. You are at Level 0. Most parts follow Way 2: products built on top of finished models. Way 1 comes up in the parts on coding assistants (Level 0) and on a safe AI-assisted coding workflow (Level 2). Way 3 comes up in Level 3, with fine-tuning and the GPU infrastructure it needs. Level 4 then shows how to run open-weight models on your own hardware.\nOur tool comparison for AI Development lists 13 tools for building, hosting and automating. Five of them fit the three ways well. Emergent fits Way 1. MindStudio and Devs.ai fit Way 2 without code. Netlify hosts web apps, and Runpod rents out GPUs for Way 3.\nEmergentApp Building \u0026amp; Testing · AI at the coreBuilds web and mobile apps from a plain-language description. MindStudioAgents \u0026amp; Automation · AI at the coreNo-code platform for building your own AI agents. Devs.aiApp Building \u0026amp; Testing · AI at the corePlatform where teams build and run AI applications and custom agents on their own data. NetlifyInfrastructure \u0026amp; Hosting · Software with AI featuresPlatform for building, deploying and hosting websites and web apps. RunpodInfrastructure \u0026amp; Hosting · AI at the coreGPU cloud for training and running AI models, billed by usage. Short descriptions from our tool comparison, based on the vendors' information — not tested by ABXK.\nFind your starting point Your first project tells you where to start. Two questions point you to a way. Before Way 3, one more test can save you a lot of work.\nTwo questions lead to your way. Test prompts and documents before you choose Way 3. What Is AI Development? Using AI vs. Building with AI (2) Your goal Your way Start with Write and understand code faster Way 1 Level 0, part 5: AI coding assistants Build an internal assistant without code Way 2, no code Level 1, part 3: a simple AI app without code Add an AI feature to your own software Way 2, with code Level 1, part 1: your first LLM API call Teach a model a house style that prompts alone cannot hold Way 3 Level 3, part 4: fine-tuning, RAG or prompting ▸Try it\nTake one AI idea from your own work, for example \u0026ldquo;answer supplier questions about our delivery terms\u0026rdquo;. Write down your answers:\nWill the finished product use a model while people work with it? Are good instructions and your documents enough, or must the model learn something new? Who could build it: someone without code, a developer or an ML engineer? Which way follows, and which part of this path will you read next? Keep the note. After Level 1, check whether your answers still hold.\n","words":1786},"date":"2026-10-14","description":"AI development means three things: using AI to write code, building apps on top of a model, and training models. Learn the roles and where to start.","image":"https://abxk.ai/images/cover/what-is-ai-development.webp","json":"https://abxk.ai/en/ai-development/learn/what-is-ai-development/index.json","kind":"page","language":"en","learning_path":{"category":"ai-development","level":0,"level_name":"Start here"},"modified":"2026-10-14","related_tools":["Emergent","MindStudio","Devs.ai","Netlify","Runpod"],"section":"ai-development","tags":["ai-development","beginner","ai-apps","no-code","fine-tuning"],"title":"What Is AI Development? Using AI vs. Building with AI","translations":[{"json":"https://abxk.ai/de/ai-development/learn/what-is-ai-development/index.json","language":"de","url":"https://abxk.ai/de/ai-development/learn/what-is-ai-development/"}],"url":"https://abxk.ai/en/ai-development/learn/what-is-ai-development/"}