👋 I’m Prakash,

In this post, I will show you how to set up a local AI chatbot using Ollama, DeepSeek-R1, and the Open WebUI. This approach offers flexibility, privacy, and control over your AI interactions without relying on cloud-based services. I’ll keep it simple and concise so we can quickly learn how to install Ollama, run LLM models like Llama 3.3 and DeepSeek-R1, and use the web interface to interact with the LLMs locally....

In this post, I will first define what LlamaIndex, RAG (Retrieval-Augmented Generation) is, describe the basic architecture pipeline that I am building, and implement the concept using Python code. LlamaIndex LlamaIndex is a powerful data framework design to connect custom(text, HTML, pdf, etc.) data sources to the Large Language Models. It acts as an interface to manage the interaction with the LLMs like loading data from source, create the index form the input data, which than used to respond to the users queries....

In this post, I will explain how to implement the Strategy Design Pattern in Node.js. To illustrate this concept, I’ll use the example of integrating multiple payment methods which can be used in the e-commerse and other applications to accept payment from customers. Whether it’s PayPal, bank transfers, or credit card. Strategy Pattern The Strategy Design Pattern is a behavioral design pattern in software engineering that enables the dynamic selection of algorithms at runtime, allowing for greater flexibility and adaptability in the application’s behavior....

In this post, I will create a small Node.js app containing the publisher and message processor, and publish it as an image to Docker Hub. Then, I will use the same image to demonstrate scaling Kubernetes Jobs using KEDA, triggered by Amazon SQS. For demonstration purposes, I will be using a Minikube local Kubernetes cluster, but in reality, this would be done in a production Kubernetes cluster Here is the architectural diagram for our workflow....

In this post, I will briefly define supervised fine-tuning of large language models (LLMs), prepare a dataset for Google Vertex AI collected from social media posts, and use Vertex AI to train Google’s LLM, Gemini to analyze the sentiment. This is a purely no-code blog post. I will demonstrate everything directly from the Google Cloud Console using Vertex AI. Introduction Fine-tuning a large language model (LLM) is the process of further training a pre-trained model to perform a specific task or work with a particular dataset....

In this post, I will define the N+1 query problem and explain how to avoid it using eager loading. Throughout my professional career, I have enhanced the response time of numerous APIs by refactoring N+1 queries. I have seen the N+1 issue in many times that use ORMs, as well as in those that use raw queries. In this blog post, I will use raw queries to demonstrate the N+1 problem and Eager Loading to solve it....

In this post, I will briefly define Large Multimodal Models (LMM), explain the data modalities of LMMs, and demonstrate Google Gemini’s multimodal capabilities using text and images. This is purely a no-code blog post. I will demonstrate everything directly from the Google Cloud Console using Vertex AI Studio and Vertex AI to showcase a simple example of using multimodal inputs (text file and image) Introduction A Large Multimodal Model (LMM) is an Artificial Intelligence (AI) model capable of processing and understanding multiple types of input data modalities simultaneously and providing outputs based on this information....

As a beginner, understanding a Kubernetes (k8s) cluster might be a little bit complicated. If you go through the k8s official documentation its overwhelming. In my personal view, the best way to learn any new tool and technology is by understanding its core concepts and architecture. In this blog post, I will try to simplify the basic concepts and common components of a Kubernetes (k8s) cluster with simple architecture diagram....

Ensuring an application runs smoothly in production is crucial. This post guides you to write and execute a simple Gatling API load test using JavaScript to ensure the stability, speed, scalability, and responsiveness of your application under a given workload. You may have written unit tests and functional tests, and the application looked good, so you shipped it to production. However, it suddenly crashes, experiences performance issues, or encounters API request timeout errors....

Logging play a very important role in software development. It helps monitoring, troubleshooting, debugging, event tracing, request tracing, security, and also can be used by Business Intelligence(BI) for reporting purpose. Logging has numerous benefits. But in this blog post, we will focus on building a real-world example to store application logs in Elasticsearch via a Node.js app by using the Winston NPM package and visualize them using Kibana. Additionally, we will dockerize the entire application....