It pains me to see all these “node” based automators charging you an arm and a leg. I can’t wait till people see the obvious and just replace them with a few prompts to Claude Code or similar.
Do these tools still make sense when you can have an agent-type setup like Claude Code or any of the competitors do that kind of low level work?
I know that vibe coding isn't quite there yet, but I wonder if low-code solutions like those are really saving any time at this point. What's the advantage if not velocity?
It is worth pointing out that both of them are workflow tools first. The AI agent part is expressed though steps where a model is invoked. This opens the door to a number of issues including vulnerabilities.
My personal, and my I say biased, opinion is that these tools do not deliver on the agentic promise. AI agents require a completely different approach.
I’m working on a self-hosted node-red-powered desktop app to connect cloud services to local resources and vice versa. Email me if you’d like to participate in the upcoming beta.
> LLMs are non-deterministic; that is each time you run something you might get a different answer.
That's a choice. With (at least) Ollama, you can set the temperature parameter so the results will be deterministic based upon the seed value. Which you can set.
It's used for things like running test cases, but there are likely other uses too.
What does it mean by "n8n" lacks a UI? All it has is a UI, what use cases does the UI not handle? The writing is very poor and does not make sense at all.
We used Node-Red at a previous job for integrating some third party software, total disaster, from upgrading, using JSON for code reviews, to overall performance.
As ever, I think each has its strengths and weaknesses, but I know which one I prefer...
In terms of 'out of the box' experience, I do agree there is space for some improved Node-RED nodes to support some of these workflows. There certainly are nodes in the community to help, but there's space for more. And that's an area I think gives Node-RED an advantage; the open source nature that allows anyone to build in the gaps of what's already there. Node-RED provides plenty of low-level building blocks to connect with pretty much anything - but having higher level abstraction nodes is needed to make it more accessible.
It's also an area we're looking at in my day job (FlowFuse CTO) where we provide a commercial platform built around Node-RED.
Hey Nick! (See my other comment someplace, which is very closely aligned to your point)
The bit I want to surface in this thread is that there is actually a lot that can be done with sub-flows and custom functions alone (even without added third-party modules) - I built my own library of AI/“agentic” subflows with raw HTTP request nodes and a simple convention for passing LLM context in payloads so I’m a bit biased here, but overall I find Node-RED superior in the amount of control I have over how data is handled.
(Also, I am running Node-RED in bun rather than NodeJS and it is way more efficient, although that means I can’t rely on some third-party nodes, but that is another story)
I have been using both for a number of years. n8n has had a resurgence recently because of its (quite clever) AI integrations (and it has higher level abstractions for doing workflows), but since I want more control I actually do my stuff (including an AÍ news summarizer and a few “agents”) in Node-RED because once you’ve sorted out your own library for things like API calls, memory, storage, etc. (which I did using just sub flows and custom JS inside them) it can be a lot simpler and more straightforward.
I see n8n as more of “citizen developer” tool (less flexibility, more built-in integrations) and Node-RED as an engineering scratchpad where you can literally build anything you want (for instance, it’s been part of my home automation setup for ages, way before Home Assistant became a thing).
I wouldn’t run _any_ of them as part of an enterprise solution, though. Security, IAM/RBAC, auditability and even general reliability would keep me up at night (even considering that Node-RED runs a surprising amount of small scale on-premises industrial automation, that’s typically in isolated networks, so I’m not factoring those in).
16 comments
[ 3.1 ms ] story [ 36.9 ms ] threadIt’s used so consistently that I’m not sure if it’s just careless editing or if it’s something new I don’t know about.
I know both make for excellent all purpose digital glue but would think that’s last resort
I know that vibe coding isn't quite there yet, but I wonder if low-code solutions like those are really saving any time at this point. What's the advantage if not velocity?
My personal, and my I say biased, opinion is that these tools do not deliver on the agentic promise. AI agents require a completely different approach.
https://github.com/node-red/node-red/blob/master/LICENSE
https://docs.n8n.io/sustainable-use-license/
We previously looked into integrating n8n but they wanted $50k for a commercial license, which didn’t make sense for us.
That's a choice. With (at least) Ollama, you can set the temperature parameter so the results will be deterministic based upon the seed value. Which you can set.
It's used for things like running test cases, but there are likely other uses too.
What does it mean by "n8n" lacks a UI? All it has is a UI, what use cases does the UI not handle? The writing is very poor and does not make sense at all.
We used Node-Red at a previous job for integrating some third party software, total disaster, from upgrading, using JSON for code reviews, to overall performance.
Its a fun toy, leave it at that.
https://github.com/activepieces/activepieces https://github.com/kestra-io/kestra
As ever, I think each has its strengths and weaknesses, but I know which one I prefer...
In terms of 'out of the box' experience, I do agree there is space for some improved Node-RED nodes to support some of these workflows. There certainly are nodes in the community to help, but there's space for more. And that's an area I think gives Node-RED an advantage; the open source nature that allows anyone to build in the gaps of what's already there. Node-RED provides plenty of low-level building blocks to connect with pretty much anything - but having higher level abstraction nodes is needed to make it more accessible.
It's also an area we're looking at in my day job (FlowFuse CTO) where we provide a commercial platform built around Node-RED.
The bit I want to surface in this thread is that there is actually a lot that can be done with sub-flows and custom functions alone (even without added third-party modules) - I built my own library of AI/“agentic” subflows with raw HTTP request nodes and a simple convention for passing LLM context in payloads so I’m a bit biased here, but overall I find Node-RED superior in the amount of control I have over how data is handled.
(Also, I am running Node-RED in bun rather than NodeJS and it is way more efficient, although that means I can’t rely on some third-party nodes, but that is another story)
I see n8n as more of “citizen developer” tool (less flexibility, more built-in integrations) and Node-RED as an engineering scratchpad where you can literally build anything you want (for instance, it’s been part of my home automation setup for ages, way before Home Assistant became a thing).
I wouldn’t run _any_ of them as part of an enterprise solution, though. Security, IAM/RBAC, auditability and even general reliability would keep me up at night (even considering that Node-RED runs a surprising amount of small scale on-premises industrial automation, that’s typically in isolated networks, so I’m not factoring those in).