Category: Uncategorized

You shouldn’t send WhatsApp marketing messages at 3 AM on Sunday. I’ll show you how to use the ‘Wait’ node and ‘Time Check’ logic in n8n to queue weekend events and release them in a batch when business opens on Monday.

By implementing a ‘business hours’ logic in your automation workflow, you can ensure that your marketing messages are delivered at the right time. This is crucial for maximizing engagement and conversion rates. In this tutorial, I’ll walk you through the steps to set up a ‘Wait’ node and ‘Time Check’ logic in n8n.

Follow along and learn how to:

• Queue weekend events and store them safely
• Release them in a batch when business opens on Monday

Don’t let your marketing messages go to waste. Implement ‘business hours’ logic in your automation workflow today.

#ShahiRaj #AutomationWorkflows #n8n

Building a dashboard is hard, but building a notification is easy. I’ll show you how I pull data from the Stripe API, Google Analytics, and my Postgres DB, format it into a clean HTML message, and push it to a Telegram Channel via a Cron trigger every Monday at 9 AM.

Want to see the code? Check out my GitHub repository for the n8n workflow.

#ShahiRaj #n8n #Automation

Handling live audio streams is scary. But, I’ll break down the architecture: Exotel (SIP) -> Whisper (Transcribe) -> OpenAI (Decide) -> n8n (Action). How to build a bot that listens and reacts in under 1.5 seconds. Check out our blog for more technical insights and get in touch if you want to build a similar setup.

Building a Voicebot that can handle live audio streams is a complex task, but it’s achievable. If you’re interested in learning more, subscribe to our newsletter and stay updated on the latest tech trends.

#ShahiRaj #n8n #VoiceBot #Technical #Exotel #Whisper #OpenAI

Traditional code breaks when a user says ‘Meet me next Tuesday after lunch.’ I’ll show you how I use Large Language Models (LLMs) to convert fuzzy natural language into strict ISO-8601 timestamps for the Google Calendar API.
With AIMeetGTBot, I can handle ambiguous inputs like ‘next week’ or ‘2 o’clock’. It’s all about contextual understanding and intent recognition. Check out the code and see how you can apply this technique in your own projects.

I got tired of SSH-ing into my server to clean logs and prune Docker images. So I wrote a Cron job and deployed Watchtower.

Here is the exact Bash script I use to keep my Debian VM clean automatically every morning at 6:25 AM:

#!/bin/bash

# Clean logs
sudo /usr/local/bin/docker-systemctl restart zabbix

# Prune Docker images
sudo docker system prune -af

# Clean up Docker volumes
sudo docker volume rm $(docker volume ls -f dangling=true -q)

# Clean up unused Docker networks
sudo docker network rm $(docker network ls -q -f is-isolated=true)

# Reboot the server
sudo shutdown -r now

With this script, I can ensure my server is always in a healthy state, even when I’m not around.

Want to learn more about automating server maintenance? Reach out to us at ShahiRaj.

ChatGPT is amazing, but it has one major limitation: it doesn’t know your business secrets. I’ll show you how I use n8n + Pinecone to create vector embeddings of private documents, allowing a Large Language Model (LLM) to answer questions based strictly on your internal files without hallucinations.

With this setup, you can connect your LLM to your private data and get accurate answers every time. No more relying on public knowledge or making assumptions. Learn how to create a secure and private knowledge base using vector databases and LLMs.

ChatGPT is amazing, but it has one major limitation: it doesn’t know your business secrets. I’ll show you how I use n8n + Pinecone to create vector embeddings of private documents, allowing a Large Language Model (LLM) to answer questions based strictly on your internal files without hallucinations.

With this setup, you can connect your LLM to your private data and get accurate answers every time. No more relying on public knowledge or making assumptions. Learn how to create a secure and private knowledge base using vector databases and LLMs.

The biggest fear with AI is that it will make things up. This is known as ‘hallucination.’ To prevent hallucination, we need to configure the AI to stick to the provided context (SOPs) and say ‘I don’t know’ if the answer isn’t in the manual. In this article, we’ll explore how to configure the ‘Temperature’ and ‘System Prompt’ in n8n to achieve this.

First, let’s understand the concept of hallucination. Hallucination occurs when the AI generates an answer that is not based on the provided context. This can happen when the AI is trained on a large dataset, but the training data is not representative of the real-world scenario.

To prevent hallucination, we need to configure the AI to stick to the provided context. This can be achieved by setting the ‘Temperature’ parameter in n8n. The ‘Temperature’ parameter controls the level of creativity of the AI. When set to a low value, the AI will provide more factual and less creative responses.

Additionally, we need to configure the ‘System Prompt’ in n8n. The ‘System Prompt’ parameter controls the context in which the AI generates responses. By setting the ‘System Prompt’ to the relevant SOP or manual, we can ensure that the AI sticks to the provided context.

By configuring the ‘Temperature’ and ‘System Prompt’ parameters in n8n, we can prevent hallucination and ensure that the AI provides accurate and reliable responses. Contact us to learn more about how to configure n8n for hallucination control.

ChatGPT is a game-changer, but it’s limited by its reliance on public data. What if you could harness the power of Large Language Models (LLMs) with your own proprietary data? That’s exactly what I’ve been experimenting with using n8n, Pinecone, and OpenAI.

The setup involves ingesting documents from Google Drive and SQL databases, creating vector embeddings using Pinecone or Qdrant, and then feeding relevant context to OpenAI. The result is an LLM that answers questions based strictly on your private files.

This approach has huge implications for companies looking to unlock the full potential of their data. By combining the capabilities of n8n, Pinecone, and OpenAI, you can:

• Create a custom knowledge graph that reflects your company’s unique data
• Develop a tailored conversational AI that speaks to your customers and employees
• Unlock new insights and revenue streams by analyzing your proprietary data

Ready to take your data to the next level? Let’s dive in and explore the possibilities.

Handling audio streams can be a real challenge when building voicebots. But what if you could set up a webhook from Exotel, process speech-to-text with Whisper, and generate a natural response in under 2 seconds? That’s exactly what I did with our recent Voice AI project.

Here’s a step-by-step guide on how I integrated Exotel with OpenAI using n8n:

• Setup the Exotel webhook to receive audio streams
• Use Whisper for speech-to-text processing
• Feed the output into an n8n workflow to generate a natural response

Want to learn more about building voicebots with n8n? Check out our resources.