PinnedAbdullah Grewal·6d agoHow AI Agents Learned to Think: The Reinforcement-Learning Recipe Behind Agentic RAG and Deep…For two years, the standard way to give a language model “search” was retrieval-augmented generation (RAG) wrapped in a hand-written agent…A response icon2A response icon2
PinnedAbdullah Grewal·Jun 22The KV Cache Wars: How MLA, PagedAttention & Sparse Attention Let LLMs Serve Millions for Pennies…Everyone obsesses over parameter counts and benchmark scores. But the quiet war that actually decides whether an LLM costs cents or dollars…
PinnedAbdullah Grewal·Jun 8How AI Finally Killed Quadratic Attention: NSA, Mamba-3, and the Architectures Making Million-Token…Self-attention has carried the transformer for nearly a decade, and quietly overcharged it the whole time. Because every token attends to…
PinnedAbdullah Grewal·May 12AI Agents Don’t Need Vector Search Anymore: Inside the Agentic Search Stack Replacing RAG in 2026In May 2025, Anthropic did something that should have been a bigger deal than it was: they took vector search out of Claude Code. The…A response icon18A response icon18
Abdullah Grewal·Jun 2AI Agent Memory in 2026: How Mem0, Letta, and Zep Cut Tokens 90% (and Rakuten Cut Errors 97%)Most AI agents are professional amnesiacs. Open a new session and they have no idea who you are, what you’re working on, or that they…A response icon2A response icon2
Abdullah Grewal·May 4Beyond Chunk-and-Embed: How LazyGraphRAG, A-RAG, and ColQwen2 Are Rewriting Retrieval-Augmented…Until recently, every RAG stack looked roughly the same: split documents into chunks, embed them, drop the vectors into a database, run…
Abdullah Grewal·Apr 28Beyond Next-Token: How Diffusion LLMs Like Mercury 2 and LLaDA Hit 1,000+ Tokens Per Second in 2026For five years, every flagship LLM has generated text the same way: one token at a time, left to right. In 2026, a new generation of…
Abdullah Grewal·Feb 17I Gave My Code to a Swarm of AI Agents and Here’s What HappenedThe Complete Guide to Claude Code’s Swarm Mode: Turn One AI Into an Entire Dev Team
Abdullah Grewal·Nov 22, 2025Antigravity: Google’s New Agent-First IDE — How It Compares to Cursor and Kiro (Deep-dive…Google Antigravity is an agent-first IDE built on Gemini 3 that gives AI agents access to editor, terminal, and browser. Read this…
Abdullah Grewal·Sep 1, 2025Google’s New AI Model “Gemini Flash 2.5 Image Nano Banana”: The Future of Image GenerationGoogle Gemini 2.5 Flash Image (Nano Banana) — Architecture, Features, Use Cases, Pros & Cons A deep dive into Google’s Gemini 2.5 Flash…
