TL;DR: N8N News for November 2026 and What It Means for Your Business
N8N news for November 2026 shows that you can self-host AI agents and automations with no per-execution fees. That lets a solo founder work like a small team. n8n GmbH turns seven on 27 November, with 208,000+ GitHub stars and a reported $2.5 billion valuation.
- Where n8n stands now: n8n now calls itself a platform for AI agents first. It lists 1,500+ connectors and 9,000+ templates, though older official pages claim 200 or 400. Before you commit, test the three or four apps your business depends on.
- When to pick n8n over Zapier or Make: Choose n8n if you run high-volume workflows, handle sensitive data, or plan to build AI agents. You can switch between OpenAI, Anthropic, Google and open-source models without rebuilding your workflows.
- Your first project: Build an inbound lead triage agent. It needs a trigger, an AI Agent node, a strict JSON prompt and a Switch node for routing. Add a human approval step and an error workflow, then test it on 20 real past messages.
- Mistakes to avoid: Giant 60-node workflows, outdated self-hosted instances, hard-coded credentials, agents acting without approval, and ignoring the fair-code license terms.
Author Violetta Bonenkamp (Mean CEO) built the entire Fe/male Switch startup game on no-code tools. Her rule is to default to no-code until you hit a hard wall. If you want the bigger picture, this guide shows how to build a startup with N8N from day one. These n8n valuation lessons explain why investors see n8n as AI agent infrastructure.
Your first workflow could be live within 14 days. The founders who start now will spend 2027 making decisions instead of copying data between tabs.
Check out other fresh startup news and trends that you might like:
N8N news for November 2026 arrives at a symbolic moment: seven years ago this month, on 27 November 2019, n8n GmbH was entered into the Berlin commercial register, according to the n8n Wikipedia entry on the company’s history. What started as a side project by a former visual-effects artist has become one of the most-starred automation repositories on GitHub. It is now the tool I see in almost every founder conversation about AI agents. I run several companies in parallel, including CADChain and Fe/male Switch, and automation is part of how I stay sane.
This roundup covers where n8n stands right now, what the numbers say, and what founders, freelancers and small business owners should actually DO with it. I am Violetta Bonenkamp, better known as Mean CEO. I built an entire startup game and online incubator on no-code tooling, so my view is practical. I will tell you where n8n saves real money, where it quietly burns your time, and the mistakes I keep seeing in founders who jump in without a plan.
Let’s break it down.
What is n8n, and why does every founder keep talking about it?
n8n (pronounced “n-eight-n”) is a workflow automation platform developed by n8n GmbH in Berlin, founded by Jan Oberhauser and released publicly in October 2019. The name is a numeronym: the “8” stands for the eight letters between the first and last “n” in “nodemation”, a blend of “node” and “automation.” You build automations on a visual canvas by connecting nodes, where each node is an app, a service, a piece of logic or an AI model. When the visual blocks are not enough, you drop into JavaScript or Python.
To avoid confusion: in this article, “node” means a single step inside an n8n workflow, and “Node.js” means the JavaScript runtime the product is built on. Also, n8n is source-available under a “fair-code” model. Fair-code means the source code is always visible and you can self-host it, while commercial use is governed by n8n’s own license terms, with Enterprise licenses available for extra features and support. Many tutorials call n8n “open source,” and that is technically loose wording you should keep in mind before building a business on top of it.
Here is a fun piece of founder trivia. Before starting the company, Oberhauser worked in film as a visual-effects artist and pipeline engineer, with credits including Pirates of the Caribbean: On Stranger Tides. Pipeline engineering in VFX is about moving assets through many tools without breaking anything. That background explains a lot about why n8n feels like a production pipeline for business data.
Quick definition for snippet hunters
- What it is: a visual, node-based workflow automation and AI agent platform.
- Who makes it: n8n GmbH, Berlin, Germany, CEO Jan Oberhauser.
- How you run it: self-hosted on your own server or as the managed n8n Cloud service.
- License model: fair-code, source-available, with paid Enterprise options.
- Main competitors: Zapier and Make, both hosted automation tools.
- Coding: optional, with JavaScript, Python and npm packages supported.
What are the biggest N8N news points for November 2026?
Rather than chase rumors, I prefer to report what is visible and verifiable. Here is the state of n8n this month, based on its public repositories, documentation and independent coverage.
- n8n now positions itself as “The Platform for AI Agents and Workflow Automation.” The official n8n GitHub repository leads with AI agents first and classic automation second. That ordering is a strategic statement.
- The connector count has climbed to 1,500+. The main repository now advertises over 1,500 app connections and 9,000+ workflow templates.
- Model flexibility is a headline promise. n8n lets you connect OpenAI, Anthropic, Google or open-source models and switch providers without rebuilding your architecture.
- Human approvals and observability are front and center. The platform pitches multi-step AI workflows with tool use, human approval steps and full visibility into each run.
- Enterprise features keep expanding. Role-based access, audit trails and support for sensitive data are listed as standard selling points for self-hosted and secure cloud setups.
- Education around n8n is exploding. DataCamp runs a dedicated course, and freeCodeCamp-style YouTube courses covering RAG agents and multi-agent systems keep appearing.
- A community ecosystem is forming around the product. Projects such as n8n-MCP, n8n-skills and n8n Workflows show up next to n8n on the n8n mirror page on SourceForge, a sign that builders are creating tools on top of the tool.
- Seven years since incorporation. The Berlin company hits its seventh registry anniversary on 27 November.
On the money side, n8n was widely reported in October 2025 to have raised a large round led by Accel at a valuation of around $2.5 billion. For a company that began as one person’s project in Berlin, that is a European success story worth paying attention to, and it tells you investors see n8n as AI agent infrastructure, far beyond a “Zapier alternative.”
What do the numbers say about n8n’s momentum?
Numbers do not lie, though they do get quoted out of context. Here are the public figures from the n8n-io organization page on GitHub and related sources.
- 208,000+ GitHub stars on the main n8n repository. That puts it in the same popularity bracket as some of the most famous developer projects on the planet.
- 61,200+ forks, meaning tens of thousands of people copied the codebase to modify, study or self-host it.
- 15,200+ followers on the n8n-io GitHub organization.
- 1,800+ stars and 4,300+ forks on the n8n documentation repository. A docs repo with more forks than stars means people actively contribute fixes and translations.
- 1,200+ stars and 687 forks on the n8n-nodes-starter template for building custom nodes.
- 1,500+ connectors and 9,000+ templates listed in the main repository.
The stat nobody mentions: the connector count depends on who you ask
Here is something I found while researching this piece. The SourceForge mirror describes n8n as having 200+ nodes. The GitHub organization profile says 400+. The main repository says 1,500+. All three are “official-ish” sources, and they disagree by a factor of seven.
The likely explanation is that older pages were never updated, and newer counts may include community nodes and generic HTTP-based connections. My lesson for founders: NEVER pick a tool based on a headline connector count. Check whether the THREE OR FOUR apps your business actually runs on have native nodes, and test them with your real data before you commit.
Why does n8n matter for entrepreneurs, freelancers and small business owners?
Small teams lose to big teams mostly on manpower. A corporation can hire three people to copy data between systems, chase invoices and summarize customer calls. A solo founder cannot. Automation platforms close part of that gap, and n8n closes it at a price point that a bootstrapped company can live with, especially when self-hosted.
According to ScalaHosting’s guide to what n8n is and how to use it, self-hosted n8n carries no per-execution fees and your data never leaves your machine. For anyone who has watched a Zapier bill grow every month as workflows multiply, that line alone explains the hype.
My take as a parallel entrepreneur
One of my operating rules is simple: “Default to no-code until you hit a hard wall.” At Fe/male Switch, we built a full startup game with quests, a token economy, an AI buddy and mentor feedback loops, all on no-code tooling. That experience taught me that early-stage founders should treat no-code and AI as their FIRST ENGINEERING TEAM. You validate the market and the mechanics before you pay developers.
n8n fits this rule better than most tools because it has an escape hatch. When the visual canvas stops being enough, you write JavaScript or Python inside the same workflow. That means the “hard wall” arrives much later than with purely visual tools. I also like that n8n treats humans as part of the loop with approval steps. I believe in human-in-the-loop AI: the machine does pattern recognition and mechanical work, and the human keeps judgment, ethics and the story.
At CADChain, our whole product philosophy is that “protection and compliance should be invisible.” Engineers should not have to become IP lawyers to stay safe. The self-hosting option in n8n follows a similar logic for data: if your customer data never leaves your server, a big chunk of your GDPR headache shrinks without anyone reading a regulation. For European founders, that is a real advantage.
n8n vs Zapier vs Make: which one should a bootstrapped founder choose?
Coverage often describes n8n as a source-available alternative to hosted tools such as Zapier and Make. Here is how I explain the choice to the founders I mentor.
Pick Zapier if…
- You want zero setup and zero server responsibility.
- Your workflows are short, linear “when X happens, do Y” chains.
- You run a low volume of tasks and the monthly bill stays small.
Pick Make if…
- You like visual branching and moderate logic without writing code.
- You are fine with a hosted service and usage-based pricing.
Pick n8n if…
- You run high-volume workflows where per-task pricing would hurt.
- You handle sensitive data (health, legal, IP, HR) and want it on your own server.
- You want to build AI agents with tool use, memory and human approvals.
- You or a freelancer on your team can write a little JavaScript or Python.
- You want to switch between OpenAI, Anthropic, Google and open-source models without rebuilding.
My provocative take: most founders pick Zapier first because it feels safe, then spend a weekend migrating to n8n six months later when the bill hurts. If you already know you will build AI agents, start with n8n on the cloud plan and move to self-hosting once your workflows are proven. You skip one painful migration.
How do you build your first n8n AI agent workflow? A step-by-step guide
First, a definition. As this Medium guide to n8n workflows and AI agents puts it, an AI agent is an LLM-powered loop (LLM meaning Large Language Model) that plans the next step, picks a tool, runs it, reads the result and repeats until the goal is reached. The same author offers a lovely metaphor: “Think of a workflow as ‘wires and switches,’ and an agent as the ‘electrician’ that decides which switch to flip next.”
Here is the process I recommend for a first project. We will build an inbound lead triage agent, because almost every business has messy inbound emails or form submissions.
- Write the job description before you open n8n. One sentence: “Read every new contact form submission, classify it as sales lead, support request, partnership or spam, and route it.” If you cannot write this sentence, you are not ready to automate.
- Choose your hosting. Start on n8n Cloud for speed, or self-host on a small virtual server if your data is sensitive. Self-hosting removes per-execution fees.
- Add a trigger node. Use a webhook from your website form, a Gmail trigger or a schedule trigger that checks a mailbox every 15 minutes.
- Add an AI Agent node. Connect your preferred model provider with your API credentials (API meaning Application Programming Interface, the way software talks to software).
- Write a strict system prompt. Tell the model to return ONLY one of four categories plus a one-line summary, in JSON format. Loose prompts create loose outputs.
- Add a Switch node for routing. Sales leads go to your CRM and a Slack channel, support goes to your helpdesk, partnerships go to your inbox, spam gets logged and dropped.
- Insert a human approval step for high-stakes actions. If the agent wants to send a reply to a potential client, route it to you for a one-click approval first.
- Add an error workflow. When any node fails, send yourself a message with the execution ID. Silent failures are the number one killer of automation trust.
- Test with 20 real past messages. Compare the agent’s classification with what you would have done. Fix the prompt until accuracy is acceptable for your context.
- Turn it on and review weekly. Check executions every Friday for the first month, then monthly.
If you prefer a guided path, DataCamp’s Introduction to Workflow Automation with n8n course assumes no prior experience. For a longer video format, the n8n Zero to Hero tutorial on YouTube covers RAG agents (Retrieval-Augmented Generation, where the AI looks up your own documents before answering) and multi-agent systems.
Five n8n workflow examples founders can copy this month
- Invoice chaser: every Monday, pull unpaid invoices from your accounting tool, draft a polite reminder with AI, and queue it for your approval.
- Content repurposing pipeline: when you publish a blog post, generate a LinkedIn post, a newsletter blurb and three short social captions, then save them as drafts.
- Customer interview summarizer: when a call transcript lands in a folder, summarize objections, feature requests and quotes into a shared research database.
- Competitor watch: check competitor pricing pages weekly, detect changes and send you a short diff.
- Freelancer client portal: when a client fills out a brief form, create the project folder, the task list, the contract draft and a welcome email in one run.
For deeper reading on agent design specifically, nexos.ai’s explainer on what an n8n AI agent can do is a good next stop.
What are the most common n8n mistakes founders make?
I have watched dozens of founders in our incubator community fall in love with automation and then fall out of love three weeks later. These are the patterns behind that heartbreak.
- Automating a process that does not exist yet. If you have never done the task manually ten times, you do not know the edge cases. Automate repetition, never guesses.
- Building one giant workflow. A 60-node monster is impossible to debug. Split it into sub-workflows with clear inputs and outputs.
- Self-hosting without updating. A self-hosted instance is YOUR responsibility. Outdated versions expose you to security issues that the vendor already fixed. Schedule updates monthly.
- Exposing your instance to the open internet with weak access control. Put it behind authentication, use strong passwords and restrict webhook endpoints.
- Letting AI agents act without approval. An agent that can email clients, issue refunds or edit your database needs a human checkpoint until you trust it. Gamification without skin in the game is useless, and automation without accountability is dangerous.
- Ignoring the license terms. Fair-code is different from permissive open source. If you plan to resell n8n-based services or embed it in a commercial product, read the license FAQ first.
- Hard-coding credentials inside Code nodes. Use n8n’s credential store instead. Screenshots of workflows get shared, and secrets leak.
- Skipping error handling. No error workflow means you discover failures when a customer complains.
- Chasing template counts. 9,000+ templates sounds amazing, but most need adjustment. Treat templates as sketches, not finished products.
- Locking yourself to one AI model. n8n makes switching providers easy. Use that. Prices and model quality shift every few months.
Is self-hosting n8n safe for a small business?
Short answer: yes, if you treat it like any other server you are responsible for. The GitHub repository highlights role-based access, audit trails and support for sensitive data. Those features help, though they do nothing if your server runs an ancient version with an open admin panel.
My checklist for founders who self-host:
- Use a reverse proxy with HTTPS and never expose the editor without login.
- Back up your workflows and credentials database daily, and test a restore once.
- Subscribe to n8n release notes and update on a fixed schedule.
- Separate production and testing instances once real money flows through your workflows.
- Log who changed what, especially if a freelancer or agency has access.
If none of this makes sense to you, use n8n Cloud. Paying for managed hosting is cheaper than a data breach, and you can move to self-hosting later.
What does most n8n coverage get wrong?
Here is why I wanted to write this piece instead of another “what is n8n” explainer. Most coverage focuses on features. Very little of it talks about behavior, and behavior is where founders win or lose.
Insight 1: “No-code for beginners” is half true
DataCamp describes n8n as accessible to people without programming experience, and for simple workflows that holds. The moment you build AI agents with branching, memory and data transformation, you need to understand JSON, APIs and basic logic, and the visual canvas will not hide that. I say this with love: n8n rewards founders who are willing to be slightly uncomfortable. My education principle applies here too: “Education must be experiential and slightly uncomfortable.” Learn by breaking things in a test instance.
Insight 2: the real moat is your workflow library
Everyone has access to the same models and the same connectors. What competitors cannot copy is the library of tested, documented workflows that encode how YOUR business runs. Treat every working workflow as a company asset. Name it clearly, document it and version it. At CADChain we think about IP all day, and your automation logic is a form of operational IP.
Insight 3: women founders and solo founders gain the most
I often say “Women do not need more inspiration; they need infrastructure.” Tools like n8n are infrastructure. A solo founder with a well-built set of n8n agents can run research, content, lead routing and admin like a team of four. That shifts who gets to build a company, and it shifts it toward people who were historically locked out of hiring budgets.
Insight 4: the ecosystem is the signal to watch
Community projects like n8n-MCP (connecting n8n to the Model Context Protocol, a standard for giving AI assistants access to tools) and n8n-skills suggest a second layer is forming: people building products on top of n8n. When a tool grows its own ecosystem, switching away gets harder, and building skills in it becomes a safer career bet for freelancers. If you are a freelancer, the FOMO is real: agencies already sell n8n setups to clients, and the early specialists set the prices.
Frequently asked questions about n8n in November 2026
Is n8n free?
The self-hosted community edition is free to run on your own server under the fair-code license, with no per-execution fees. n8n Cloud and Enterprise licenses are paid.
Is n8n open source?
n8n describes itself as fair-code and source-available. The code is public and self-hostable, while commercial use follows n8n’s own license terms, so check them before reselling.
Which AI models work with n8n?
OpenAI, Anthropic, Google and open-source models are all supported, and you can switch providers without changing your workflow architecture.
Who founded n8n and where is it based?
Jan Oberhauser founded n8n in Berlin, Germany. The project launched publicly in October 2019, and n8n GmbH was registered on 27 November 2019.
Can a non-technical founder use n8n?
Yes, for simple workflows. For AI agents and data-heavy automations, expect to learn JSON basics and some logic, or pair up with a technical freelancer for the first builds.
What should you do next?
The N8N news this November boils down to one picture: a Berlin company celebrating seven years, sitting on more than 208,000 GitHub stars, pitching itself as the home for AI agents, and growing an ecosystem of community tools around it. For founders, freelancers and small business owners, that means the tool is mature enough to trust with real work and young enough that early specialists still have an edge.
My advice as Mean CEO is to treat your startup like a strategic game. You are collecting information, assets and relationships faster than competitors, and automation is how a small team plays above its weight. Next steps:
- This week: list five tasks you repeat more than three times a week.
- Next week: sign up for n8n Cloud or spin up a self-hosted test instance via the official n8n website.
- Within 14 days: build the lead triage agent from the guide above, with a human approval step.
- Within 30 days: add error handling, document the workflow and decide whether to self-host.
- Every month: review executions, update your instance and retire workflows nobody uses.
Start small, keep a human in the loop, and build a workflow library that becomes part of your company’s value. The founders who do this in 2026 will spend 2027 making decisions while everyone else is still copying data between tabs.
People Also Ask:
Can you provide an overview of n8n?
n8n is a source-available, node-based workflow automation platform built by n8n GmbH, a company based in Berlin. It connects apps, APIs, and databases through a visual editor where each step in a process is a block called a node. Non-technical users can drag and drop nodes to build automations. Developers can add custom JavaScript or Python code or make direct HTTP requests when they need more control. You can run it on your own server for free or pay for the hosted n8n Cloud version.
How do you pronounce n8n AI?
n8n is pronounced "n-eight-n." The name is a numeronym for "nodemation," a blend of "node" and "automation." The "8" stands for the eight letters between the first and last "n" in nodemation, following the same pattern as abbreviations like "i18n" for internationalization.
Is n8n outdated now?
No, n8n is still actively developed and widely used. Some online discussions ask whether it is "dead" or "still relevant" as AI agents and coding assistants grow more capable. The platform keeps adding features, including native AI agent nodes, connections to large language models, and support for tools like Claude that can build and run n8n workflows. Its large community, template library, and frequent updates keep it a strong choice for automation.
Is n8n cost money?
n8n can be free or paid, depending on how you use it. The self-hosted Community Edition is free to run on your own computer or server through Docker or npm, with no per-execution limits. You only pay for your own hosting. n8n Cloud, the managed version run by the company, is a paid subscription with tiered monthly plans. Business and Enterprise plans add features like single sign-on and advanced user permissions. Check the n8n website for current pricing, since plans change over time.
What is n8n used for?
n8n is used to automate repetitive tasks across different apps and services. Common uses include syncing data between a CRM and a spreadsheet, sending Slack alerts when a form is submitted, processing incoming emails with AI, generating reports, managing leads, and building customer support chatbots. Teams also use it to move data between databases, schedule recurring jobs, and connect internal tools that have no built-in link to each other.
What are n8n workflows?
An n8n workflow is a series of connected nodes that runs a process automatically. Every workflow starts with a trigger node, such as a new email, a webhook call, a form submission, or a set schedule. The following nodes then carry out actions like reading data, transforming it, applying logic, calling an AI model, or sending results to another app. Workflows can include branches, loops, and error handling, so they can manage simple one-step tasks or long multi-step processes.
What apps can n8n connect with?
n8n offers hundreds of ready-made nodes for popular apps and services, including Google Sheets, Gmail, Slack, Notion, Airtable, HubSpot, Salesforce, GitHub, Telegram, OpenAI, and many SQL and NoSQL databases. If an app lacks a dedicated node, you can still connect to it through the HTTP Request node using its API. The community also builds and shares custom nodes, which widens the list of supported services even more.
Is n8n open source?
n8n is often called open source, but it is more accurately described as "source-available" or "fair-code." Its source code is public on GitHub, and you can view, modify, and self-host it for free. The Sustainable Use License places limits on commercial use, such as reselling n8n as a hosted service to others. For most individuals and businesses automating their own work, it functions much like an open-source tool.
How is n8n different from Zapier?
Both tools connect apps and automate tasks, but they differ in a few ways. n8n can be self-hosted, which keeps your data on your own servers, while Zapier is only available as a hosted service. n8n gives developers more room to write custom code, build complex logic, and call any API directly. Pricing also differs: Zapier charges per task, while n8n Cloud charges per workflow execution, and the self-hosted version has no execution fees. Zapier tends to be quicker for beginners, while n8n offers more flexibility and control.
Can you build AI agents with n8n?
Yes, n8n has native support for building AI agents and AI-driven workflows. It includes nodes for connecting to large language models like OpenAI, Anthropic, and Google Gemini, plus tools for memory, vector databases, and retrieval-augmented generation (RAG). You can create agents that read messages, make decisions, call external tools, and route data based on the results. This makes it a popular choice for building chatbots, AI assistants, and automated research or support systems without writing an entire application from scratch.
FAQ on N8N News and AI Workflow Automation in November 2026
What can founders learn from n8n's 2025 valuation jump?
n8n reportedly reached a $2 billion valuation in July 2025 after more than 12 term sheets competed for the deal. Later reports put it near $2.5 billion. The lessons for founders are to solve painful enterprise problems, time launches to market trends and build community before monetizing. Read 7 lessons from the n8n valuation surge
How do I protect self-hosted n8n from malicious community nodes and package attacks?
Treat every community node as third-party code. Install only well-maintained packages, pin versions, test updates in staging and give each credential minimal permissions. Keep backups and assign a named owner to each workflow, because one compromised dependency can expose customer data or interrupt cash flow. See n8n security signals for founders
Can n8n automate SEO blog publishing without hurting content quality?
Yes, if you follow the rule "only automate what you can do in your sleep." Do keyword research, tune prompts and build a template article manually first. Then automate drafts and check outputs daily. Publishing 10 to 20 interlinked subtopic articles builds topical authority. Follow the AI marketing automation workshop playbook
Where can first-time founders practice automation before touching real customer data?
Practice in a sandbox, never on live systems. Fe/male Switch offers a SANDBOX and the PlayPal AI co-founder for rehearsing startup tasks safely. The stakes are worth it: workflow automation has been linked to productivity gains of around 35%, and small businesses report saving up to 20 hours weekly. Start with N8N for newbies in 10 steps
How should a small team scale from one n8n workflow to company-wide automation?
Prove one measurable workflow first, such as hours saved on lead routing. Then turn the repeatable patterns into reusable sub-workflows that sales, support and finance can share. These patterns include approval steps, error alerts and naming conventions. Standardization keeps automations maintainable as more people build them. See how n8n works as business infrastructure
How do I calculate whether an AI automation is worth building?
Multiply minutes per task by weekly frequency. Then compare the annual hours saved against build time, model API costs and ongoing maintenance. If payback takes longer than three months, simplify or skip the automation. Prioritize frequent, rule-based and error-prone tasks like data entry and reporting. Explore AI automations for startups
Can I run n8n with local AI models so no data reaches external providers?
Yes. Self-host n8n and connect it to open-source models running on your own hardware, which keeps prompts and documents in-house. n8n publishes a self-hosted AI starter kit to speed up setup. Expect higher server costs and possibly lower output quality than top hosted models. Browse n8n's official GitHub repositories
How do I build a custom n8n node for an app without a native integration?
Start with the HTTP Request node, which handles most REST APIs. If you call that API often, build a custom node from n8n's starter template. You can then publish it as a community package so others can install it and improve it. Read ScalaHosting's guide to extending n8n
When should I use an AI agent instead of a fixed n8n workflow?
Use fixed workflows when steps are predictable, like syncing invoices. Choose an agent when inputs vary and the next action depends on context, like handling mixed customer requests. Agents cost more per run and behave less predictably, so the best setup often combines both: deterministic workflow rails with agent decisions inside. Understand what an n8n AI agent can do
How do I give an n8n agent knowledge of my company documents?
Build a RAG pipeline. Split your documents into chunks, store their embeddings in a vector database and let the agent retrieve relevant passages before answering. Start with a narrow source such as your FAQ or pricing docs. Then test answer accuracy against real customer questions. Watch the n8n Zero to Hero RAG tutorial
