About KAVRIQ

KAVRIQ is a technical publication exploring maths, software systems, and AI.

Much of today’s AI content focuses on tools, demos, and surface-level abstractions. KAVRIQ takes a different approach: explaining the mathematics, models, and systems behind modern AI from first principles.

The goal is not simply to keep up with the latest framework, but to understand how things work underneath: the probability, optimization, representation, reasoning, and engineering ideas that shape AI systems.

About Me

I am Ravi Shankar, a software engineer working across AI systems, developer platforms, and cloud infrastructure.

I graduated in Computer Science from MNNIT Allahabad and have worked across startups, product engineering, cloud infrastructure, and large-scale platform environments. My experience includes THB, Urban Company, Flynote, Trinkerr, Oracle, and Salesforce.

KAVRIQ is where I document what I am learning, building, and trying to understand. My current areas of study include machine learning, deep learning, agentic AI, retrieval, evaluation, AI infrastructure, and the mathematical ideas behind these systems.

I care deeply about clarity, depth, mathematical intuition, practical judgment, and the distance between an impressive demonstration and dependable engineering.

Why KAVRIQ Exists

I am the son of a teacher, and teaching has always been one of my passions.

That interest has taken different forms over the years: mentoring colleagues, writing explanations, discussing ideas, and opening a small coaching center in Greater Noida while working full-time in Gurugram. Across each of them, I found the same satisfaction in taking something difficult, understanding it carefully, and helping someone else see it more clearly.

KAVRIQ is a continuation of that instinct. It gives me a place to learn deeply, organize what I understand, and teach it through first-principles explanations. Writing an article forces me to examine gaps in my own thinking; conversations and podcasts let me learn from others while making their knowledge useful to a wider audience.

Teaching and learning give KAVRIQ its form. Curiosity gives it direction. The aim is to make difficult ideas approachable without stripping away the depth that makes them valuable.

KAVRIQ exists because teaching is work I find meaningful—and because the best way I know to understand something is to try to explain it well.

A Working Philosophy

Understanding how things work is more valuable than chasing tools.

The goal is not to keep up with trends, but to build a mental model that lasts: mathematical where needed, practical where useful, and honest about uncertainty.