AI

Once upon a time (Feb 2001, around 25 years ago), Agile manifesto arrived as the superhero in cape for software development: promising speed, adaptability, and a razor-sharp focus on what customers actually wanted. While Agile’s soul was anti-framework, yet, in the winding corridors of enterprises, marred by the fear of chaos, the fear of losing control, its cape have been snagged on the rigid hooks of frameworks like Scrum and SAFe, transforming Agile into what? a “veiled Waterfall” ? ...

How Decision tree works, the basics - split, gini, entropy, information gain

Mult Agent Orchestrated Blog writer using Google ADK, that learns and matches users writing style

Ever wonder how your email app just knows what spam is? Or how Netflix uncannily suggests the exact B-movie you were secretly in the mood for? The answer is Machine Learning (ML). Before ML, software was a bit of a bureaucrat. You had to write explicit, rigid if-this-then-that rules for every single possibility. This works fine until the real world, in all its messy glory, shows up. ML flips the script. Instead of feeding a machine rules, you feed it examples. The machine’s job is to look at thousands or millions of examples and figure out the patterns on its own. ...

Product Managers (PMs) used to treat AI like a distant, complex cousin – interesting if you were building the AI itself, but mostly irrelevant to the daily grind. Now, nearly every PM is wrestling with GenAI tools, and a surprising number are even trying to build their own. This isn’t just a trend; it’s a fundamental shift that’s re-engineering what it means to be a PM, all thanks to AI’s knack for demolishing the tedious, repetitive tasks that sucked the life out of our days. ...

Building a Simple Analytics Agent with MotherDuck + OpenAI Agents SDK

Part 1 of “AI ain’t Magic, It’s Math” Before AI became what it is today, machines behaved like bureaucrats: they only did what you explicitly told them to do. Software was ==Rules Based==. Meaning: you wrote conditions, and the machine reacted to those conditions. First you create the rules mentally If A happens then X, If B and C happends then do Y, If A and D and E Happens then Do Z Then hard code it ...

..coming soon

Turning chaotic inboxes into structured, automated order & inquiry workflows

You hear these two words thrown around all the time, and they’re almost always used together. But here’s the thing: they do two completely different jobs. Getting them right is the foundation of all digital security. So what’s the deal, exactly? Authentication is the bouncer at the club door checking your ID. Its only job is to answer one question: “Who are you?” Authorization is the VIP wristband you get after you’re inside. It tells the bouncer which rooms you’re allowed into. Its job is to answer: “What are you allowed to do?” Authentication always comes first. You can’t figure out what someone is allowed to do until you know who they are. That’s it. One proves your identity, the other checks your permissions. Let’s break down the most common ways apps do both. ...