@ai-architecture

Ahmed Adawy

@ai-architectureCape Town, South AfricaJoined July 2026

Computational Nuclear Engineer • Scientific Python • AI Engineering

About

Scientific Python • AI Engineering • Numerical Methods • Building practical tutorials and open-source tools.

Available for

Nothing here yet.

Ahmed Adawy's blogs

The Architecture of Thought Blogadawy.hashnode.dev17 posts

Articles Comments3

Recently published

AAAhmed Adawyadawy.hashnode.devSep 13 · 6 min read

The Guardian Algorithms: Why Applied Mathematics Is the Ultimate Shield in Modern Cyber Defense

The Guardian Algorithms: Why Applied Mathematics Is the Ultimate Shield in Modern Cyber Defense Introduction: The Illusion of Perimeter Defense In modern enterprise security, a common misconception p

AAAhmed Adawyadawy.hashnode.devSep 11 · 4 min read

The Illusion of Autonomous AI Agents: Why Context Drift and State Pollution Are Killing Your Production Backends

The developer community is currently gripped by agentic fever. Frameworks like CrewAI, LangGraph, and AutoGen promise a future where autonomous LLM swarms break down complex human objectives, call ext

AAAhmed Adawyadawy.hashnode.devSep 4 · 6 min read

Why Your Async Python Services Crash Under AI Load (And How to Build a Real Production Inference Pipeline)

When a software engineer builds a prototype for an AI microservice using FastAPI, asyncio, and PyTorch or Hugging Face, everything runs smoothly on a local environment. But as soon as the service hits

AAAhmed Adawyadawy.hashnode.devAug 30 · 5 min read

Architecting High-Throughput LLM Pipelines: Resolving Memory Drift, GIL Contention, and Async Bottlenecks in Production

Transitioning a Generative AI pipeline or Large Language Model (LLM) service from an experimental Jupyter Notebook to a mission-critical, high-throughput production environment introduces performance

AAAhmed Adawyadawy.hashnode.devAug 28 · 4 min read

Beyond the Hype: The Fundamental Math Behind Next-Token Prediction in LLMs

Generative AI often looks like magic from the outside. You feed a prompt into a Large Language Model, and it seamlessly generates structured code, translates complex texts, or engages in multi-turn r