Archive - Into AI

I'm opening a few consulting slots

If you're interested, just reply to this email.

Jul 7 • Dr. Ashish Bamania

Arithmetic Intensity, Simply Explained

A no-jargon breakdown of Arithmetic Intensity and the Roofline model, and how they are used to optimize LLM inference.

Jul 6 • Dr. Ashish Bamania

Your company can pay for Into AI (here's how)

A ready-to-send email template to expense your subscription.

Jul 4 • Dr. Ashish Bamania

This Week In AI Research (21-30 June 26) 🗓️

The top 10 research papers of this week: GPT‑5.6, Sonnet 5, Meta's real‑time brain‑to‑text decoder, a 35B model that beats trillion‑parameter LLMs …

Jul 3 • Dr. Ashish Bamania

June 2026

NVIDIA GPU Ecosystem, Simply Explained

A guide to NVIDIA GPU architecture, interconnects, and scaling in plain English.

Jun 30 • Dr. Ashish Bamania

This Week In AI Research (14-20 June 26) 🗓️

The top 10 AI research papers that you must know about this week.

Jun 24 • Dr. Ashish Bamania

A hardware-level tour of how LLMs generate text

Understand in depth how LLM Inference actually works at the CPU and GPU level.

Jun 22 • Dr. Ashish Bamania

This Week In AI Research (7-13 June 26) 🗓️

The top 10 AI research papers that you must know about this week.

Jun 15 • Dr. Ashish Bamania

Build and train a Diffusion LLM from scratch

An end-to-end guide to training a LLaDA-style Diffusion LLM and using it to generate text.

Jun 14 • Dr. Ashish Bamania

Latent Mixture-of-Experts (Latent MoE), Clearly Explained

A lesson on NVIDIA's Latent Mixture-of-Experts (MoE) architecture that powers the Nemotron-3 Super and Ultra models.

Jun 12 • Dr. Ashish Bamania

I'm not writing another cringe "Fable 5" post

And instead, teaching other "uncool" and useful concepts that actually matter.

Jun 11 • Dr. Ashish Bamania

How LLMs are Actually Trained

In the last lesson, we learned how the Transformer architecture powers an LLM.

Published on AlgoMaster Newsletter • Jun 11

#nojs-banner { position: fixed; bottom: 0; left: 0; padding: 16px 16px 16px 32px; width: 100%; box-sizing: border-box; background: red; color: white; font-family: -apple-system, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; font-size: 13px; line-height: 13px; } #nojs-banner a { color: inherit; text-decoration: underline; } This site requires JavaScript to run correctly. Please turn on JavaScript or unblock scripts