Author: Shabbir Poonawala

I’m excited to share my experience building @MisriCalendarBot, a Telegram bot that provides instant weather and prayer times for any city globally. As a developer, I wanted to create a bot that would be both efficient and accurate.

I chose to use AWS to power the bot, leveraging their speed and scalability to ensure that users receive the most up-to-date information. I also integrated Official Mumineen.org timings to guarantee the accuracy of our data.

One of the key features of the bot is its ability to handle user queries and provide personalized responses. I achieved this using a combination of natural language processing (NLP) and machine learning algorithms.

Overall, building @MisriCalendarBot was a rewarding experience that allowed me to push the boundaries of what is possible with Telegram bots.

In this post, I’ll walk you through how I built an automated lead funnel using n8n, Vector DBs, and Webhooks.

The goal was to create a system that could capture leads while the user is celebrating New Year’s Eve. To achieve this, I used n8n to connect to a Vector DB, which acts as a real-time database. When a new lead is generated, n8n triggers a Webhook that sends the data to a marketing automation tool. The result is an automated lead funnel that’s ready to go from day one.

Want to learn more about building an automated lead funnel with n8n and Vector DBs? Check out our blog for more technical insights and best practices.

I just spent the last 72 hours migrating our entire core to AWS, and it was a wild ride. But it was worth it – our new infrastructure is lightning-fast, super secure, and boasts 99.99% uptime.

As a developer, I know how crucial it is to have a solid tech foundation. But what does that mean for business owners? It means having a scalable, reliable, and efficient system that can handle the demands of the AI era.

We’re using n8n to create custom workflows and Vector DBs for blazing-fast data retrieval. And with Webhooks, we’re ensuring seamless integrations.

We’re opening 3 slots for Workflow Audits this week. Let’s identify and resolve your bottlenecks before Jan 1st. Reach out to schedule a call!

Over a weekend, we moved our production VM shahiraj (Debian/Docker) from GCP to its new AWS home. The goal was to take advantage of AWS Bedrock and AWS SageMaker for building even more complex RAG agents. We achieved this by parallel processing in the @MisriCalendarBot and our E-com Tracking bots, which now feel snappier on AWS.

Our cloud spend is now optimized, keeping our SaaS prices competitive for our clients. This migration has been a game-changer for ShahiRaj Automation, and we’re excited to see the impact it will have on our clients’ businesses.

If you’re interested in learning more about our migration story and how we’re using AWS to take our AI automation capabilities to the next level, check out the ShahiRaj Blog.

As a developer, I’ve had my fair share of frustration with traditional staffing methods. That’s why I decided to build a Commercial AI Agent using n8n, Vector DBs, and webhooks.

The goal was to create a system that could handle customer interactions with minimal human intervention. With n8n as the workflow automation engine, I integrated Vector DBs for efficient data storage and webhooks for seamless communication.

Now, our AI Agent can process and respond to customer queries in real-time, eliminating the need for extensive training or manual intervention. Shahi Raj can help you build a similar system for your organization.

Want to learn more about the tech stack behind our AI Agent? Reach out to schedule a consultation.

#ShahiRaj #AIDevelopment #N8n #VectorDBs #Webhooks

I recently built a scalable call center solution using n8n, Vector DBs, and our AI Voice Agents. The goal was to create a system that could handle an unlimited number of concurrent calls.

The solution involved using n8n to handle webhooks and API integrations, Vector DBs to store and manage call data, and our AI Voice Agents to answer and qualify calls in real-time.

One of the key challenges was ensuring that the system could scale horizontally. I used n8n’s workflow capabilities to distribute the load across multiple nodes, and Vector DBs to handle the high volume of call data.

The result was a system that could handle 50 concurrent calls without breaking a sweat. And the best part? It was all built using open-source tools and our AI Voice Agents. Learn more about our AI-powered solutions and discover how you can build a scalable call center that meets your business needs.

#ShahiRaj #CallCenter #n8n #VectorDBs

We’ve built a ‘Sales Copilot’ RAG agent to supercharge our sales team’s productivity. In this post, I’ll dive into the technical architecture and implementation details of our Sales Copilot.

Our Sales Copilot integrates with our Vector DBs, leveraging n8n workflows and Webhooks to provide instant, approved responses to technical questions. This enables our junior reps to ask the bot and get the perfect answer instantly.

I’ll walk you through the code logic and architecture behind our Sales Copilot, highlighting the key technologies and integrations that make it work.

Stay tuned for the next part of this series, where I’ll share more about the business impact of our Sales Copilot.

#n8n #VectorDB #SalesCopilot #ShahiRaj

I’ve seen it happen: CEOs pasting sensitive client contracts into standard ChatGPT to get a summary. Stop. You’re potentially training a public model on your private data. Here’s a better approach: Self-Hosted RAG Agents. We spin up a private Vector Database (like Pinecone or pgvector) on your own server. The AI ‘reads’ your data, but never ‘learns’ from it for the public. It lives in a walled garden. You get the intelligence of LLMs without the data leak risk.

vectorDb = Pinecone.create_index('my_index')
rag = n8n.create_workflow('my_workflow')

I recently built a Q&A bot using n8n and Vector DBs for a client. The goal was to create a system that could scan a large number of documents and provide accurate answers to employee queries. Here’s a high-level overview of how I approached the project.

First, I set up a webhook to receive incoming queries from employees. Then, I used n8n to process the query and retrieve the relevant documents from the Vector DB. Finally, I implemented a simple logic to extract the answer from the document and return it along with the page number.

It was a fun project, and I’m excited to see the impact it will have on the client’s internal knowledge management.

Want to learn more about building a Q&A bot? Check out our blog for more technical insights.

#n8n #VectorDBs #QandA

SaaS apps trap your data in their ecosystem, making it difficult to integrate with other tools. But what if you could have a custom bot that connects your ecosystem and gives you a command line for your life? That’s exactly what we do at Shahi Raj.

We build a middleware layer that takes a simple Telegram text and routes it to Google Sheets, Trello, or SQL automatically. It’s not just a chat, it’s a command line for your life. Learn how we can help you build your own second brain.

Here’s an example of how it works:

• Forward invoice to bot: bot saves it to Drive
• Tell bot 'Remind me': bot adds task to Calendar
• Search for PDF: bot finds it in seconds

Want to know more? Let’s chat about how we can help you build your own second brain.

#ShahiRaj #SecondBrain #Python