NEWSLETTER
Why build things in the difficult way when you can design them intelligently?
As a data chain data scientist, I have explored several frames such as Langchain and Langgraph to build ai agents using Python.
The previous illustration is an article that I wrote at the end of 2023, entitled “Taking advantage of LLMS with Langchain for supply chain analysis, a GPT -driven control tower.”
At that time, I was exploring how to use Langchain to build an agent acting as a control chain control tower.
A year later, I discovered the power of the low N8N code platform to build the same type of solution in just a few clicks.
In this article, we will explore how to easily build ai agents to automate the analytics workflows of the supply chain using N8N.
We will also see how to implement the same AFo Tower Agent that I originally built with Langchain 18 months ago, now using only one low code.
Agent for supply chain control towers using Langchain
My first IA automation project project using N8N was for a client who wanted a control chain control tower equipped with a chat interface.
A supply chain control tower It is a set of panelsARDS and reports connected to warehouse and transport management systems that use data to monitor critical events throughout the supply chain.
In a previous article published in Toing Data Science, I experienced Langchain to connect a control tower to an ai agent.
The idea was to build a planning and execution agent that
- Process the user -written user application
- Generate the appropriate SQL query
- See the database and store the results
- Formulate a clear simple response
After several iterations, I found the structure of the correct chain and the indications to offer precise results.
The solution worked well because he had already gained experience using Langchain and other frames to build ai agents.
How are we supposed to maintain this complex configuration?
However, to offer this as a service, I needed tools that made the solution easier to implement, maintain and improve, even with a limited knowledge of Python.
It was then that I discovered N8N.
We are going to immerse ourselves in that in the next section.
Agent for supply chain control towers: built with N8N
What is N8N?
N8N is an open source workflow automation tool that allows you to easily connect applications (email, CRMs, messaging systems), API and ai models as Langchain.
Build workflows connecting pre -constructed nodes.
For example, the previous workflow Process emails
- The first node collects emails from a Gmail account.
- The email content and metadata are sent to the ai Agent node, which extracts the relevant information.
- The third node processes the exit using Javascript.
- The final node loads the results on a Google sheet.
Code was not needed to build this workflow, except the third node, which uses only two JavaScript lines.
Since I work with a team of supply chain consultants who have limited skills from Python, this was a change of play for me while I was looking to develop my service offer.
They can use it, adapt and maintain this workflow after a brief N8N training session.
ai Supply Chain Control Tower N8N Workflow
The workflow of the control tower of the ai supply chain is a bit more complex, but even simpler than its Python version.
Includes two sub-works.
The main sub-workflow includes a chat interface and the ai agent.
For ai agent node, you need
- Connect a LLM (chat model) using a node where your API credentials enter
- Add a memory node to administer conversation
- Add a tool node for the SQL consultation, linked to the second sub-workflow
ai ai generates a SQL query and sends it to the “call consultation tool”, which executes the consultation.
The sub-workflow includes a code node that cleans the consultation (eliminate additional spaces and block risky commands such as Delet).
The output is sent to a Large nodethat executes the consultation and returns the results.
The process is very soft and requires a limited configuration:
This configuration does not require Python skills and my consultants can manage directly.
The results are comparable to those of the Python version.
To obtain step -by -step configuration instructions, see my YouTube tutorial
Conclusion
This example shows how easy it is to replicate an ai agent built with python, using N8N and minimum code.
Does that mean that Python is no longer necessary for supply chain analysis? Not definitely!
Like many low code platforms, the characteristics are limited to what is available within the framework.
That is why I use it as a complement to my analytical products.
To do that, you can use the HTTP application node to connect your workflow to your analysis backend.
What else? Easy connectivity for many services.
Another reason why I chose N8N to enrich my analytical products is how easy it is to add additional connections.
For example, if you want to add a lazy interface or record conversations to a Google sheet, simply add a new node to your workflow.
If you are starting your N8N trip and needs inspiration, Feel free to explore my templates.
About me
We are going to connect LinkedIn and <a target="_blank" href="https://twitter.com/Samir_Saci_” rel=”noreferrer noopener” target=”_blank”>twitter; I am an engine chain engineer who uses data analysis to improve logistics operations and reduce costs.
To consult or advice on sustainable analysis and transformation of the supply chain, do not hesitate to contact me through Logigreen consulting.
Samir satisfying | Data science and productivity
A technical blog that focuses on data science, personal productivity, automation, operations research and sustainable …samirsaci.com
<script async src="//platform.twitter.com/widgets.js” charset=”utf-8″> (Tagstotranslate) Agent ai (T) Automation (T) Data Science (T) Logistics
