Articles
In-depth research, comprehensive guides, and long-form thoughts on modern technology.
How to Install OpenAI Codex CLI: A Step-by-Step Guide
OpenAI's Codex CLI brings autonomous coding to the terminal. This guide covers Node.js setup, npm installation, authentication, and first commands across macOS, Linux, and Windows.
How to Build a Simple AI Web Scraper With Python
This guide walks through building an AI web scraper in Python that fetches a page, cleans HTML, converts it to Markdown, and answers user queries with a small language model.
Running SQL Concurrently Across Three Remote DuckDB Servers With Quack
DuckDB's experimental Quack protocol enables concurrent SQL across remote instances. A test on three EC2 nodes shows sub-second coordination via barrier-synchronized threads.
Designing a Persistent Knowledge Layer That Refuses to Guess
A Microsoft MVP proposes a hybrid knowledge architecture that compiles understanding into durable artifacts instead of reconstructing answers from chunks on every query.
Pinpointing Failure in LLM Multi-Agent Systems Is Harder Than It Looks
A PSU and Duke team built the first benchmark for automated failure attribution in multi-agent LLM systems, and even top reasoning models struggle to find the culprit.
Scaling Laws, Carefully: How to Fit and Interpret AI Models
Scaling laws predict how AI performance improves with more compute. However, fitting them is tricky. This article explores the pitfalls and nuances behind the curves.
Harness Engineering: The Key to Recursive Self-Improvement in AI
The next leap in AI might not be bigger models, but better harnesses. This article explores how engineering the system around the model can unlock recursive self-improvement.
AI Agents: Architecture, Frameworks, and the Future of Intelligent Systems
A comprehensive overview of AI agents. This guide covers the core architecture, planning, tool use, evaluation, and frameworks defining the future of autonomous AI.
Controlling Reasoning Effort in LLMs: Does More Compute Mean Better Output?
The cost of inference is a key barrier to AI adoption. This article explores methods for controlling the reasoning effort in LLMs to optimize for cost and quality.
Building an AI Text Detector From Scratch: Techniques and Challenges
Building an AI text detector from scratch requires a mix of statistical analysis and machine learning. This article breaks down the core components and common pitfalls.
LWiAI Podcast #252: GPT 5.6, Grok 4.5, and the Future of AI in 2040
This episode recap explores the predictions for AI in 2040, including the performance of GPT 5.6 and the unique capabilities of Grok 4.5.
LWiAI Podcast #253: Opus 5, Gemini 3.6, and the New AI Frontier
A deeper dive into the latest AI model releases discussed on LWiAI Podcast #253, including a performance analysis of Opus 5 and Gemini 3.6.