Swapan Rajdev

The Only Real Moat in AI: Speed

Why Speed Is the Only Moat Left in AI and Startups

Jul 25 • 

Swapan

LLM Evaluation Metrics: Your Production Scorecard

Evaluation Metrics to measure for Enterprise Success

Jul 17 • 

Swapan

5

May 2025

Multi-Layered Guardrails for AI Agents

Runtime Protection Alongside Model Alignment Is Essent

May 1 • 

Swapan

March 2025

A Summary of LLM Post-Training Techniques

Why fine-tuning, reinforcement learning, and test-time scaling are essential to turning language models into useful agents.

Mar 27 • 

Swapan

3

Building Moats in the Age of AI

Strategies for Sustainable Competitive Advantage

Mar 19 • 

Swapan

5

2

February 2025

The AI Stack: Understanding the Key Layers Powering the AI Ecosystem

A Breakdown of AI Infrastructure, Data, Models, Tooling, and Applications for Builders and Businesses

Feb 26 • 

Swapan

2

January 2025

Understanding GPU Components for LLMs

In the fast-evolving world of artificial intelligence, each new GPU release is met with a mix of excitement and scrutiny.

Jan 8 • 

Swapan

DeepSeek-V3: The Pinnacle of Open-Source AI

DeepSeek-V3 currently stands as the best open-source AI model, decisively outperforming competitors such as Llama 3.1 405B, Qwen, and Mistral in…

Jan 3 • 

Swapan

November 2024

Understanding Agentic AI Architecture

A Comprehensive Guide for Beginners to Experts

Nov 26, 2024 • 

Swapan

3