Effortless AI: No-Code Automation Using N8n Cloud and OpenAI Vision API | Hacker News Reader

Effortless AI: No-Code Automation Using N8n Cloud and OpenAI Vision API

n8n-automation.com

50 pointscrazytest15 comments

Even after clicking around the product pages a little, I’m afraid I still don’t understand what an n8n is or what it’s seeking to automate. Is “n8n” a colloquialism amongst front-end people, like “i18n” or “a11y”? And as to “automation,” I’m guessing from context that this is like a Zapier kind of product?

And that the present announcement is that this product has an integration with OpenAI Vision API now, provided you’re willing to store your API keys in the open?

I’m surprised to see it characterized as “no-code” but then require HTTP headers to be entered by hand, and a valid JSON payload to be hand-built to send over to the relevant OpenAI API.

The broader concept, though, seems cool. I can imagine the sort of visual, business-side-person kind of power user who might have a lot of fun playing with this type of workflow tool.

n8n is the tool that is used to build the automation shown in the tutorial, leveraging the OpenAI Vision API. That is possible because it offers quite an open way to make HTTP Requests, using the node for it. Yes, you need to define a valid JSON as an input for the API, but I would still consider it way more beginner-friendly than using the SDKs where you would have to write "real" code and care about deployment and executions.

n8n is a direct Zapier competitor with source available ("fair code license"), https://news.ycombinator.com/user?id=janober is the founder.

Lots of interesting ways to extend workflow systems involving LLMs and interfaces to them.

https://docs.n8n.io/choose-n8n/faircode-license/

https://hn.algolia.com/?q=n8n

Automate what? What's n8n? And how no-code/ effortless is it if I'm deciding between base64 encoded images vs url images, REST vs GraphQL API, etc. in the first few paragraphs with no background info on how to make those choices?

Thanks for the feedback. I was not sure in what detail I should go, but decided to name the options and make an easy choice to go for if you follow along the tutorial. If you then decide you want to dive deeper, you can dig into the documentation of e.g. OpenAI. I definitely should link to for example the article where I go into detail about HTTP Requests and what they are, but I could not fit it into the article itself.

Do I need to read the entire article to find out what the tutorial is actually aiming to achieve? That's not very helpful to the beginner it's supposed to serve...

Thanks for the valuable feedback. Just added a paragraph under the head picture to make clear what it is exactly about. Is this helpful or would you do it in another way?

So, this is interesting, but much more interesting is the langchain stuff N8N has added to their latest release. Makes it seamless to add in an army of interns to your workflow.

https://n8n.io/langchain/

You're right. The Langchain integration is great, but you need an extra account for it. But it is actually in the backlog to write a tutorial about whats possible with Langchain too.

fyi, a separate account should not be necessary any more as the functionality got integrated into the regular version and is so available for all users by default.

great news! haven't seen it as I had to update my workspace manually, but just did it.

I'm not quite sure how n8n works after reading the docs, but our company Reflect provides an AI-driven approach to automation that may be similar, but for a more narrower use-case: automated end-to-end testing.

You can see a video example of how it works in our docs here (https://reflect.run/docs/recording-tests/testing-with-ai/), but the idea is that you describe the actions and assertions you want to take in plain-text prompts, and the AI interprets those prompts in real time and executes them against a running browser session. In practice, it's a lot like writing a manual test script and having it automatically execute. We use both GPT 3.5 and 4 and will be releasing Vision support once OpenAI has deemed the gpt-4-turbo-with-vision model ready for production use.

actually, that is what it is about, that it is not a narrower use-case, and you can extend and change the workflows however you like them. but an interesting way to present your own company. Seems like you are doing something similar to what https://www.octomind.dev/ is doing, right?

what are some useful use cases startup entrepreneurs?

I guess there are a lot. If you build a great UI you can use a workflow like this to score presentation slides or whatever you like

Reply on news.ycombinator.com