Author: Shabbir Poonawala

I recently worked on a project to build an AI All-in-One Widget for Shahi Raj. The goal was to create a widget that could assist visitors instantly and empower the sales team with pre-qualified leads. In this post, I’ll take you through the technical aspects of building this widget.

The widget is powered by a combination of n8n, Vector DBs, and Webhooks. We used n8n to integrate with various APIs, Vector DBs to store and manage leads, and Webhooks to trigger automated workflows. The backend is built on AWS Mumbai, ensuring a fast and reliable experience for our users.

In terms of code logic, we used a combination of JavaScript and TypeScript to develop the widget. We also leveraged various libraries and frameworks to simplify the development process.

Want to learn more about the technical aspects of building the AI All-in-One Widget? Check out our documentation for more details.

In this post, I’ll walk you through how to build a Custom Telegram Utility Bot using n8n and PostgreSQL. This bot will act as a central hub for your operations, enabling your team to pull data and trigger workflows with a simple command.

Here’s a high-level overview of the architecture:

• n8n as the workflow engine
• PostgreSQL as the data store
• Telegram as the communication channel

In this example, we’ll use the n8n Telegram webhook to receive updates from the bot and trigger workflows accordingly. We’ll also use the PostgreSQL driver in n8n to pull data from the database.

Here’s some sample code to get you started:

const n8n = require('n8n');const pg = require('pg');const telegram = require('node-telegram-bot-api');const bot = new telegram('YOUR_TOKEN');

Want to learn more about building Custom Telegram Utility Bots? Check out our resources page for more information.

As a developer, I recently built an AI Doctor Appointment Agent using n8n, Vector DBs, and Webhooks. The goal was to create a scalable solution that could handle a high volume of patient bookings and rescheduling requests.

Here’s a high-level overview of the architecture:

• n8n: Used as the workflow automation engine to handle patient bookings, rescheduling, and cancellations.
• Vector DBs: Utilized as the data storage solution for patient information and appointment schedules.
• Webhooks: Used to receive and send notifications between n8n and the front desk staff’s digital calendar.

The entire system is hosted on AWS, ensuring a reliable and scalable infrastructure. The AI-powered appointment agent is able to sync perfectly with the digital calendar, ensuring the team always has an accurate view of the day.

Want to learn more about building a scalable AI Doctor Appointment Agent? Reach out to us for a consultation.

I recently built a Digital Chief of Staff using n8n, Vector DBs, and Webhooks. The goal was to create an AI assistant that could pull research, track finances, and manage calendars directly via Telegram.

I used n8n’s workflow editor to design the logic for the bot. The workflows are triggered by Webhooks, which are then processed by a Node.js script. The script interacts with the Vector DB to store and retrieve data.

The bot is hosted on our AWS Mumbai server, which provides a scalable and secure environment for the AI assistant to run in.

Want to learn more about my tech stack and architecture? Check out our website for more details.

I recently worked on a project where we built a custom n8n workflow to automate tasks for a hospital. The goal was to create a ‘Silent Partner’ that would handle the heavy lifting, freeing up staff to focus on more strategic work.

We connected existing tools (e.g., WhatsApp, CRM, Sheets) so data flows effortlessly without manual intervention. The workflow was then deployed to AWS Mumbai, ensuring it’s always online and processing tasks 24/7.

The tech stack used includes n8n, Vector DBs, and Webhooks. The code logic is designed to handle various scenarios and exceptions, providing a robust and scalable solution.

One of the key challenges was migrating 400+ workflows to the new system. We used a combination of scripts and manual configurations to ensure a smooth transition.

Want to learn more about building your own Silent Partner? DM me, and let’s discuss your project.

I recently built an Always-Ready FAQ Assistant using n8n, Vector DBs, and Webhooks. The goal was to create a 24/7 digital concierge for websites that can instantly assist customers with FAQs.

The workflow involves pulling data from the knowledge base, filtering out repetitive queries, and powering the AI-powered support widget with our AWS Mumbai backend.

Check out the tech stack and architecture behind our Always-Ready FAQ Assistant.

#n8n #VectorDBs #ShahiRaj

When I set out to build the ShahiRaj Data Collector, I knew I had to choose the right tools for the job. I decided to use n8n as my workflow automation platform, which seamlessly integrates with AWS.

The Android APK is built using standard Java and Kotlin libraries, with a focus on ease of use and minimal setup for field workers.

For validation and data processing, I leveraged the power of Vector DBs and Webhooks to ensure instant visibility and minimize errors.

Want to see the code? Check out the ShahiRaj Data Collector repository on GitHub.

#ShahiRaj

I’ve always been fascinated by the idea of automating administrative tasks for sales teams. Recently, I built an AI-powered Appointment Widget using n8n, Vector DBs, and Webhooks. Let me walk you through the tech stack and architecture.

The widget uses n8n as the workflow engine, which integrates with Vector DBs for real-time availability checks and Google Calendar syncing. Webhooks enable seamless communication between the widget and the sales team’s calendar. The AWS Mumbai server hosts the widget, ensuring optimal performance and data security.

One of the key challenges was ensuring the widget’s availability and accuracy. I employed a combination of natural language processing and machine learning algorithms to gather necessary context before the meeting. By doing so, the sales team is always prepared and empowered.

Check out the ShahiRaj blog for more insights on AI-powered solutions and how they can transform your business.

As a developer, have you ever struggled with the ‘where is that file?’ problem? You know, when a new employee or even a senior manager needs to find a specific policy, technical spec, or past project detail, and they waste 30 minutes searching or asking you.

I recently worked on a project to build Custom RAG (Retrieval-Augmented Generation) AI Agents at ShahiRaj. This isn’t just ChatGPT; it’s a private, secure AI trained exclusively on your business data.

I vectorized the documents using PostgreSQL/pgvector database on AWS and implemented a web widget for instant answers. Our RAG Agents cite the exact document and page number, and because they only look at your data, they don’t make things up.

Imagine giving every employee a personal assistant that has memorized every document your company has ever produced. That is the efficiency of a RAG Agent. Read more about our Custom RAG AI Agents and how they can benefit your business: https://www.shahiraj.com/custom-rag-ai-agents

#ShahiRaj #CustomRAG #AIAgents

I’m excited to share the technical details behind our MisriCalendarBot build, powered by the Mumineen.org API and hosted on our high-speed AWS Mumbai server.

We leveraged Node.js, n8n, and Vector DBs to create a seamless user experience for Hijri dates, Miqaats, holidays, and Namaz timings.

Stay tuned for more technical insights into our bot’s architecture and implementation. Learn more about our tech stack and expertise.