Blog posts

ML from Scratch Episode 2: All about Data

12 minute read

Published: July 13, 2026

Data is the single most important element in Machine Learning. It is the one ingredient you absolutely need to get right for the entire recipe to turn out well. So before we dive headfirst into any algorithm, it is crucial to understand data better—because it is the foundational bedrock of any ML objective. Understanding data means understanding the underlying patterns it carries. And often, this is not quite straightforward. Different forms of data hide different kinds of patterns, and it is our job to format, process, and transform them accordingly so that our algorithms can actually understand, interpret, and learn these patterns.

ML from Scratch Episode 1: Math Behind Machine Learning and Gradient Descent Algorithm

12 minute read

Published: July 09, 2026

I got the greatest opportunity to revisit ML once again from scratch, and I want to utilize this chance to do it right. I want to begin this journey at the right place, and that would be the “Foundations—the Math behind Machine Learning.” Most learners are reluctant to step into this zone, not because it is difficult, but because of an uncomfortable lie we spread amongst ourselves: ML Math is hard. But well to be honest, it ain’t easy and I won’t deny that, but it is also a straightforward truth that it is not impossible. With the right material and the right tutorial, even the math behind ML can become a comfortable concept. In this blog post, I aim to outline the necessary math concepts that serve as prerequisites for ML. I will give a very generic introduction, but the deep dive and practice are something I leave up to you.

Prompt Engineering vs RAG vs Finetuning: Strategic AI Customization guide

10 minute read

Published: June 24, 2025

In today’s rapidly evolving AI landscape, off-the-shelf large language models (LLMs) often fall short when faced with specialized business requirements. While these foundation models possess remarkable general capabilities, they frequently struggle with domain-specific terminology, proprietary data contexts, and unique organizational needs. This performance gap has catalyzed three powerful customization approaches: Prompt Engineering, Retrieval-Augmented Generation (RAG), and Fine-Tuning. Each method offers distinct advantages for transforming generic AI into a precision instrument for specialized tasks.

Decoding MCP: A comparison between Model Context Protocol vs Rest API

4 minute read

Published: June 04, 2025

The AI Isolation Problem: Why MCP Was Born

Picture a brilliant consultant locked in a windowless room. No internet, no documents, no tools—just raw intelligence. This was the reality of AI systems before MCP. Despite their astonishing capabilities, large language models (LLMs) remained trapped in silos, disconnected from the databases, APIs, and tools that could make them truly useful. Every new integration—whether fetching live data from PostgreSQL or automating Blender 3D modeling—required custom code, special prompting, and fragile plumbing. Developers faced an N×M integration nightmare: N AI models needing bespoke connections to M data sources.

Beyond ChatGPT: How Block Diffusion Bridges the Gap in Language Modeling

3 minute read

Published: May 29, 2025

As a researcher who’s spent years wrestling with language model limitations, I still remember my frustration when ChatGPT choked on generating coherent text for my queries. That fundamental tension—between the creativity of diffusion models and the precision of autoregressive architectures—has haunted our field. Until now. The breakthrough work in “Block Diffusion” (Arriola et al., ICLR 2025) isn’t just another incremental improvement—it’s the architectural bridge we’ve desperately needed. Let me walk you through why this paper could be an interesting direction for the future of Language Modeling.

Collective Transport: Engineering without Blue Print

7 minute read

Published: May 13, 2025

Ant colonies routinely achieve the remarkable feat of transporting objects far exceeding individual capacity—from hefty food items to nesting materials—often navigating complex and cluttered terrains. This stands in stark contrast to coordinated human efforts, which often rely on explicit planning and communication and can falter under similar constraints. The ants’ success hinges not on a pre-designed plan, but on sophisticated, decentralized strategies emerging from local interactions.

The Poor Man’s Finetuning Duel: A comprehensive report on LLM fine tuning on Llama and DeepSeek

7 minute read

Published: March 05, 2025

Drawing on extensive (and often frustrating) experience with CUDA memory limitations, I approached the prevalent claims of “efficient” 7B model fine-tuning with skepticism. Benchmarking Llama-2-7B and DeepSeek-7B under strictly controlled conditions (single A100, 40GB VRAM) yielded results that were profoundly surprising. They exposed a significant gap between the rhetoric of efficiency and its practical reality in constrained environments. This analysis fundamentally altered my perspective; allow me to detail this critical reality check.

Unlocking the Magic of ChatGPT: A Journey Through Transformer-Based Language Models

4 minute read

Published: December 22, 2024

Imagine an AI that crafts poetry, debugs code, and explains quantum physics—all while adapting to your unique conversational style. This isn’t science fiction; it’s the reality of ChatGPT, a pinnacle achievement in modern AI. But how does it really work? The secret lies in a revolutionary architecture called the Transformer. In this deep dive, we’ll demystify Transformers, build a Shakespeare-generating model from scratch, and reveal what powers ChatGPT’s linguistic brilliance.