Building AI Task Suggestions: Teaching a Machine to Read My Emails

I built an AI system that reads my emails and meeting notes to suggest tasks automatically. Here's what I learned about Gmail APIs, LLMs, and why processing 50 emails at once seemed like a good idea.

WebMCP - A Builder’s Tutorial for Advanced Newcomers

Understanding Anthropic Agent Skills: From Tool-Calling to Expertise Modules

Agent Skills transform AI agents from generalists that need constant instruction into specialists with reusable expertise. This article demystifies what Skills are, how they differ from tools and prompts, and why they represent a fundamental shift in how we build reliable AI systems.

Creating Text Embeddings Using Gemini (Python)

A practical guide to generating text embeddings using Google's Gemini embedding model. This covers API usage, batch processing, chunking strategies, and the invariants that matter for production embedding pipelines.

Authentication and Authorization in Backend Systems: Understanding the Flow

Authentication and authorization are the gatekeepers of modern applications. This article explains how these security mechanisms work, why they're different, and how to implement them correctly in backend systems.

Building a JIRA Chatbot: The AI-Powered Way

I spent two weeks building a JIRA chatbot that answers any question in plain English. Here's what I learned about combining JIRA's REST API withGemini, the mistakes I made, and what actually works in production.

Connecting JIRA to Our Task Manager: The Messy Reality

I spent two weeks connecting JIRA to our internal task manager. Here's what I learned about JIRA's REST API, the mistakes I made, and what actually worked in production.

Why Data Engineering Exists (and Why Data Ingestion Comes First)

Data engineering didn’t start with tools. It started with broken systems. This article explains why data engineering exists, how ingestion became its foundation, and why correctness begins at the moment data enters a system.

Building AI Agents from Scratch

Learn how to build AI agents from scratch using Python, explicit tools, and a transparent control loop for safe, reliable reasoning.

Automating Steel QC: A Complete Guide to Vision-Based Inspection Setup

A practical guide for steel manufacturers looking to automate quality inspection. Learn how to set up cameras, lighting, data pipelines, and AI models in harsh industrial environments.

AI Agents: What They Are, How to Use Them, and When Not To

Large Language Models excel at processing and generating language, but tackling multi-step problems requires AI agents. Learn what they are, how to build them, and when to skip them.

Building Bulletproof File Upload Systems: Backend Security & Validation

Deep dive into the critical backend engineering decisions behind secure file upload systems. Learn why every uploaded file is a potential threat and how to build defensive systems that protect your infrastructure while maintaining performance.

Conquering Document Parsing: Mastering PDFs, DOCX, and the Chaos of Real-World Files

Master the art of parsing chaotic real-world documents: why every PDF is a potential disaster, how to build systems that expect failure, and battle-tested strategies for extracting meaning from the messiest files.

What Makes FastAPI Fast?

An accessible, step-by-step explanation of why FastAPI is fast, how it works under the hood, and what makes it different from older Python web frameworks.

Pydantic AI Gateway: Unified Interface for Multi-Provider AI Access

A comprehensive guide to Pydantic AI Gateway, covering its architecture, key features, provider management, and how it simplifies multi-provider AI access with built-in observability and cost management.

Activation Functions: The Neurons' Decision Makers

A comprehensive guide to activation functions in neural networks, exploring their mathematics, advantages, and disadvantages

Loss Functions: Measuring Model Performance

A comprehensive guide to the most important loss functions in deep learning, exploring their mathematics, use cases, and implementation strategies

PyTorch Deep Learning Workflow: From Data to Deployment

A comprehensive step-by-step guide to building, training, and deploying deep learning models with PyTorch

'LLM as a judge' for evaluation using Pydantic AI

A comprehensive guide to using LLMJudge in Pydantic AI for evaluating subjective qualities of LLM outputs based on rubrics.

LLM-as-a-Judge: What It Is, How to Use It, and When Not To

A practical guide to using LLMs as evaluators: when to use them, when not to, and how to design effective rubrics for robust, reliable AI evaluation.

Building My First Web App with FastHTML: What I Learned

An in-depth guide to FastHTML, covering its architecture, core concepts, component system, and how it revolutionizes web development with pure Python.

EXAMPLE - Linear Regression, Done Properly

Linear regression is the first model everyone uses and the first model most people misunderstand. This is a full technical article that treats linear regression as a bounded system - invariants first, paper execution before code, failure modes before confidence, and a mastery gate at the end.

How to Write Good Technical Articles

Using AI to churn out text is trivial. Turning an average article into a solid technical piece is not. A strong article proves that the author has actually learned the subject, not just described it. It takes the reader from confusion to clarity, sets the right boundaries, exposes failures and edge cases, and leaves them with the confidence that they can execute the idea themselves. You are not writing on an armchair writing. You are producing an artifact for Divami, meant to percolate skill across the organization.

How to Write with AI

In the age of AI where 99 out of 100 articles are being written by AI, it is very critical to differentiate yourself from the rest of the pack with quality and authenticity.

Convolutional Neural Networks (CNNs): A Deep Dive

An in-depth guide to Convolutional Neural Networks (CNNs), covering their architecture, key concepts, popular models, and real-world applications.

Why Shrinking Models Makes Them More Powerful

Quantization, pruning, and student–teacher training reveal a core truth of modern AI intelligence is resilient, redundant, and far less dependent on precision and scale than we once believed.

Mastering File Upload and Storage in Data Engineering: A Complete Guide

A comprehensive deep dive into modern file upload and storage systems, from handling the 5 V's of data to implementing secure, scalable upload mechanisms and choosing the right storage solutions for your data engineering needs.

Building an AI-Powered Query System on Salesforce Objects

A deep dive into Salesforce objects, their structure, and how to leverage LLMs to build an AI-powered querying system for dynamic business insights.

Why Talking Matters in the Age of AI

Free, unedited speech is not a shortcut - it is an engine. Speech accelerates recall, expands associative range, reveals hidden intent patterns, and produces the raw material from which high-quality reasoning emerges.

Understanding Salesforce Objects: A Comprehensive Guide

A comprehensive guide to Salesforce objects, standard and custom object types, how they structure data, and practical examples with code for extracting object metadata.

Understanding Embeddings: The Semantic Backbone of LLMs

Embeddings are the semantic backbone of LLMs, transforming raw text into vectors that machines can understand. This article explores how embeddings evolved from simple statistical methods to the sophisticated contextual representations that power modern AI.