Blog

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.

I recently built a scalable AI e-commerce tracker using n8n, Vector DBs, and Webhooks. The goal was to create a system that could handle 800+ monthly updates on our AWS Mumbai server without breaking a sweat.

The system connects directly to our Shipmozo/BigShip accounts, triggering an automated update via WhatsApp the moment a status changes (Dispatched, Out for Delivery, Delivered). We’ve optimized the workflow to minimize latency and ensure seamless updates.

The tech stack includes:

• n8n for workflow automation
• Vector DBs for real-time data storage and retrieval
• Webhooks for instant updates

Check out the ShahiRaj blog for more insights into our e-commerce automation solutions.

This post is part of our Developer Log, where we share technical expertise and behind-the-scenes stories.

I recently had the opportunity to architect and deploy a production-grade digital workforce using n8n and AWS. The result was a seamless migration of our 400+ n8n workflows and AI agents to a dedicated AWS server in Mumbai.

Read the full story on how we achieved near-zero latency, reliability, and security for our WhatsApp and Voice bots.

I’d love to share more about the tech stack and architecture behind this project. DM me if you’re interested in learning more about custom n8n workflows and AWS infrastructure.

As a developer, I’m excited to share how we built the Ai Lead Gen Widget using n8n and AWS. Our goal was to create a 24/7 sales agent that could answer complex questions and qualify leads instantly.

We started by setting up a n8n workflow that integrates with our AWS-hosted n8n instance. The workflow uses Webhooks to push data directly to your CRM or WhatsApp.

On the frontend, we developed a user-friendly interface that allows users to interact with the sales agent. We used Vector DBs to store user data and n8n’s built-in logic to handle complex questions and lead qualification.

Want to learn more about our development process? Reach out to me and I’ll be happy to share more details.

I recently built an AI Meeting Scheduler for ShahiRaj using n8n and Vector DBs. Here’s a high-level overview of the architecture.

The bot lives in WhatsApp and Telegram, and when a user wants to meet, it uses the n8n workflow to check the user’s Google Calendar in real-time.

The Vector DBs are used to store the user’s availability, and the bot uses this data to offer only the slots the user wants to show.

The bot then handles the confirmation and sends the invite—all inside the chat window.

Since we migrated to our AWS Mumbai backbone, the response time is near-instant.

Check out the code and architecture on my GitHub repository: https://github.com/your-username/ai-meeting-scheduler