Context engineering is your only durable AI edge
Context engineering means filling an AI model’s context window with your own domain knowledge, using structured graphs, deterministic tools, and persistent memory instead of simple RAG.
Tag: Backend
Machine learning engineering lessons from a short February
February was short but it made a few things obvious: solo work stalls ML projects, code is not documentation, and an MLOps stack that ignores the deployment environment never ships. Notes from a month of ML and backend work.
Coding Pong in Python: why OOP beats vibe coding
Coding Pong in Python is a solid way to practice OOP: inheritance for the ball, a compound check instead of distance() for paddle collisions, and a scoreboard class kept separate from the game logic.
WooCommerce 10.5.3 patches a critical Store API bug
WooCommerce 10.5.3 patches a path validation bug in the Store API batch endpoint that let attackers skip nonce checks and gain admin access via CSRF. Don’t roll back to 10.5.2; update now and test your checkout.
DAX Filtering is killing your Power BI performance
Ahmad Wael breaks down why overusing FILTER() in DAX forces work into Power BI’s slow, single-threaded Formula Engine, and how a simple predicate lets the Storage Engine (VertiPaq) handle it instead.
CSS exploit in Chrome: how CVE-2026-2441 actually works
Ahmad Wael explains CVE-2026-2441, a Use-After-Free bug in Chrome’s Blink CSS engine triggered through @font-feature-values. It’s not literal code execution via CSS, but it’s a real reason to update your browser now.
7 reasons WordPress.com site ownership beats the DIY stack
After 14 years of stitching together WordPress hosting stacks, here is why I moved to WordPress.com: managed infrastructure, 99.999% uptime, real developer tools, and backups that actually restore.
Scaling feature engineering pipelines with Feast and Ray
Most feature pipelines are just Python scripts and CSV files until they break in production. This post covers using Feast for feature management and Ray for distributed compute to fix training-serving skew and cut latency.
Agentic RAG caching: reducing latency and token waste at scale
Agentic RAG setups waste time and money answering the same questions repeatedly. This post covers a two-tier semantic and retrieval cache, plus the validation logic needed to keep it from serving stale answers.
Production data architecture for real AI systems
Most AI features never leave the notebook. Notes on the production data architecture that gets them into a live site: separated layers, database integrity over raw insert speed, and the state and permission problems agents bring with them.