At some point, every team building with LLMs hits the same fork in the road: do you get better results by writing a smarter prompt, or by actually retraining the model on your own data? It sounds like a technical detail, but the choice affects your cost, your timeline, and …

Generative AI has gone from research demo to daily habit faster than almost any technology before it. It writes our emails, drafts our code, and designs our slides. But every capability that makes it powerful also opens a new question about trust, cost, and control. Understanding both sides isn't optional …

A few years ago, "prompt engineer" sounded like a job title from the future. Now models write their own prompts, refine them, and even critique each other's outputs. So the question worth asking isn't whether prompting still works — it's whether the skill of prompting still matters once the AI gets …

For years, AI meant a chatbot that answered your questions and waited for the next one. That's changing fast. Agentic AI systems don't just respond — they plan, act, and adapt on their own to get a job done. If you've ever wondered what happens when AI stops waiting for instructions …

Introduction

In today's session, I learned about Retrieval-Augmented Generation (RAG) and AI Agents using the Haystack framework. The session focused on how modern AI systems can retrieve information, use external tools, reason through problems, and generate more accurate responses.

Retrieval-Augmented Generation (RAG)

RAG combines information retrieval with Large Language Models …

Introduction

In this session, I built an AI-powered Notes application by integrating a local Large Language Model with FastAPI and MongoDB.

This project goes beyond basic APIs by combining model inference, backend logic, and persistent storage — similar to how real-world AI systems operate.

The goal was to create a system …

Introduction

Day 6 of my GenAI Learning Challenge.

Today, I integrated a local LLM with FastAPI to build a usable AI backend. Instead of running the model separately, user input is sent through an API and processed by llama.cpp.

This is a key step toward building real AI applications …

Introduction

In Generative AI systems, most models are accessed through APIs. However, running models locally is becoming increasingly important for privacy, cost efficiency, and offline usage.

To understand how local inference works, I ran a Large Language Model (LLM) on my system using llama.cpp.

In this session, I focused …

Introduction

In Generative AI (GenAI), building models is only one part of the process. The real foundation lies in how we handle and structure data before it reaches the model.

To get started, I focused on three essential Python concepts:

• Variables
• Lists
• Dictionaries

These form the backbone of how data …