# Build an AI product with an independent developer > An AI product uses a model to do a useful job for your users, such as answering questions about their documents or drafting text. To build one, you define the task, test the model on real examples first and wrap it in a reliable product. ErdiBuilds is an open marketplace where clients post app, SaaS, website and game projects, see which independent developers from the EU fit best based on what they have already built, and pay per milestone. Canonical: https://erdibuilds.app/build/ai-product Last updated: 2026-11-06 (November 6, 2026) ## What an AI product is, and is not A large language model (A program trained on lots of text that can read, write and answer in natural language, such as the models behind ChatGPT and Claude.) is a building block, not a product. The product is the job around it: what goes in, what comes out, who checks it and how it fits into someone's day. People build assistants that answer from company documents, tools that summarise calls, helpers that draft emails or listings, systems that sort and tag incoming messages, and extraction tools that turn invoices or forms into data. The frame around the model is usually a [web app](https://erdibuilds.app/build/web-app) or a [SaaS](https://erdibuilds.app/build/saas). ## Typical building blocks [Find My Stack](https://findmystack.app/?project_type=ai-product) can suggest tools for your case. Common parts are these: - A model provider through an API (A set of rules and addresses that lets one program ask another for data or actions.), such as OpenAI or Anthropic. See [OpenAI developers](https://erdibuilds.app/hire/openai). - Prompts (The instructions and examples you send to a model to shape its answer.) that are written, tested and versioned like code. - Retrieval augmented generation (A pattern where the system first finds relevant passages in your own data and gives them to the model, so answers rest on your content.) when answers must come from your documents. - A vector database (A database that stores text as numbers so it can find passages with a similar meaning.) such as the pgvector extension in Postgres. - Evaluations (A set of example questions with expected answers that you run after each change to see if quality went up or down.) to measure quality instead of guessing. - Guardrails, usage limits and logging, so a single user cannot run up costs or see another person's data. [Python](https://erdibuilds.app/hire/python) and [Node.js](https://erdibuilds.app/hire/nodejs) are the usual back end languages. > Models and their prices change often. Ask your developer to keep the model choice in one place of the code, so switching later is easy. ## An example plan in milestones Start with proof that the model can do the job. Each milestone is funded before work starts and runs 1, 2, 3, 4, 6 or 8 weeks. Milestones are from 50 to 2,000 each in your project currency. 1. For example, week 1 to 2, feasibility: collect twenty or so real examples, try models and prompts, report what works and what does not. You decide to continue with facts in hand. 2. For example, weeks 3 to 6, working product: the app around the model, accounts, saving results, a first retrieval setup if needed. 3. For example, weeks 7 to 8, quality and safety: evaluation set, handling of wrong or empty answers, rate limits, privacy review of what is sent to the provider. 4. For example, weeks 9 to 10, launch: monitoring of cost and errors, feedback buttons for users, handover. The weeks are for illustration. Developers set the length of each milestone in their proposal, and each real milestone runs 1, 2, 3, 4, 6 or 8 weeks. Cut the first one short on purpose: its main output is a decision. ## What belongs in the project description - The task in one sentence and who benefits. - Examples of good inputs and the answers you would be happy with. These become the test set. - Your data: what exists, in which format and how sensitive it is. - How wrong answers should be handled and who reviews critical results. - Privacy needs, for example no personal data sent to outside providers. - Expected number of users and how often they use the product, since model use is billed by volume. - Whether you prefer a specific model or provider. If personal data is involved, data protection rules may apply. This is general information, not legal or tax advice. ## How to recognise a suitable developer Be wary of anyone who promises perfect answers. A good developer talks about testing, fallbacks and limits, shows past work with models in real products and explains costs per request in simple terms. They should be happy to build the evaluation set with you. On ErdiBuilds, each proposal carries a match score (A number from 0 to 100 that shows how well a developer fits your project, with plain reasons such as shared technologies, similar past work, budget fit and availability.). A developer needs at least 40 to apply, and 80 or more gets a Top match badge. Look at the Expert in list for AI related technologies. Developers are independent professionals, not our employees. ## Frequently asked questions ### Which AI model should I use? It depends on the task, speed, cost and privacy. A good developer tests two or three models on your own examples in the first milestone and shows the results. ### Can the product answer from our own documents? Yes, with retrieval: the system finds the right passages first and the model answers from them. Quality depends on how clean and well structured your documents are. ### What if the AI gets something wrong? Models can make mistakes. Plan for it with sources shown next to answers, a feedback button, limits on risky actions and human review where mistakes matter. ### Are there running costs after launch? Yes. Model providers charge by use, and hosting has its own cost. Ask your developer to estimate usage costs for your expected volume and to add a spending limit. ### Who owns the prompts and code? Unless both sides agree otherwise in writing on the platform, the rights to the result of a milestone pass to the client once that milestone is paid in full. Ask that prompts and test sets are part of the repository you receive. ## Related - [Hire an OpenAI and LLM developer](https://erdibuilds.app/hire/openai): Add chat, writing or analysis features with AI models. What an LLM developer does, what to ask, how it is planned and how matching works. - [Hire a Python developer for data and AI](https://erdibuilds.app/hire/python): Python powers servers, data work and AI features. See what a Python developer builds, what to ask them and how a project is planned. - [Build a SaaS with an independent developer](https://erdibuilds.app/build/saas): Plan a SaaS product: accounts, subscriptions, technologies, an example milestone plan, what to write in the brief and how to pick a developer. - [Build a web app with an independent developer](https://erdibuilds.app/build/web-app): How to plan a browser-based web app: common technologies, an example milestone plan, what to write in the brief and how to judge a developer. - [What Is an MVP? A Plain Guide for Founders](https://erdibuilds.app/guides/what-is-an-mvp): An MVP is the smallest version of a product that real users can try. Learn what it includes, what it leaves out and how to plan one. - [Hire a Developer to Build Something New](https://erdibuilds.app/help/build-new): Turn an idea into a working product. See how to prepare, how milestones structure the work and what to ask developers before you start.