Articles
AI-assisted engineering guides
Guides and working patterns for teams building with AI coding tools.
Latest article / 01GitHub vs. GitLab for AI Development in 2026: The Features That MatterCompare GitHub Copilot and GitLab Duo across coding agents, review, CI, security, customization, models, self-hosting, and cost before choosing a platform.
Latest article / 02How to Make AI Videos: Two API Workflows and the Real Cost of a 15-Second ReelBuild a repeatable 15-second vertical reel with Runway or Google Veo, assemble it in a template, and budget for generated seconds, retries, and rendering.
From Thousands to Millions of Tokens: Do Context Windows Still Need Optimization?A source-backed history of LLM context windows from GPT-3 to million-token models—and a practical guide to deciding when to send everything, retrieve, or summarize.
How to Use REA with Claude Code and Codex: Inspect, Rebuild, and VerifyConnect REA to Claude Code or Codex, trace an Electron export, verify recovered behavior, and turn the evidence into a tested implementation.
How to Use Local LLMs for Coding: Setup, Workflows, and Honest LimitsA practical guide to local coding models: choose a workflow and memory budget, connect Ollama to Aider, test changes on your own repo, and understand the privacy and quality limits.
Can AI Agents Test Your App Like a Human? OpenAI’s New Computer Use ExplainedOpenAI added a hosted browser to the Agents API. Learn how agents can exercise web-app flows, what evidence to collect, and where automated tests and human approval still matter.
Graph Engineering for AI Coding Agents: Graphify, Claude Code, and CodexUnderstand code knowledge graphs and agent workflow graphs, then use Graphify to explore a repository in Claude Code or Codex with evidence-backed examples.
Motion Design with Claude Code: Prompts That Produce Better Web AnimationsA practical prompt workflow for better web animation in Claude Code, with copy-paste recipes for reveals, feedback, and state transitions plus accessibility and performance checks.
How to Evaluate a Coding Agent: Traces, Tests, Reverts, and Cost per Merged PRA source-backed framework for evaluating coding agents on real repository work, from traces and tests to post-merge rework and fully loaded cost.
Build a PR Risk Gate with Jev: Typed AI Decisions for Safer Code ReviewsA practical TypeScript guide to classifying pull-request risk with Jev while deterministic policy and human reviewers retain control.
Can Jev Help With Programming? Where TypeSafe’s Decision Model Fits in AI Coding WorkflowsA source-backed guide to TypeSafe AI's Jev: where a fast typed decision model can help coding agents, where it cannot, and how to test it safely.
Claude Code vs. OpenAI Codex vs. Cursor: Which AI Coding Agent Should You Use for the Rest of 2026?A source-backed late-2026 comparison of Claude Code, OpenAI Codex, and Cursor across workflow, context, autonomy, security, team controls, and cost.
AI Coding Agent Security: Sandboxes, Permissions, Prompt Injection, and Safe AutonomyA practical security guide to Claude Code and Codex: sandboxing, permissions, prompt injection, secrets, MCP risk, safe autonomy, and team rollout.
Context Engineering for Claude Code and Codex: A Practical GuideA practical context-engineering workflow for preventing context rot in Claude Code and Codex with better instructions, delegation, compaction, and handoffs.
MCP vs. Agent Skills vs. Plugins vs. Custom Agents: What Should You Use?A practical guide to choosing MCP, Agent Skills, Plugins, and Custom Agents in Codex and Claude Code—and combining them without unnecessary complexity.
Is Code Review Still Worth It in the Age of AI? Best Practices for Human and AI ReviewA risk-based guide to modern code review: what AI should inspect, where human judgment still matters, and how to review AI-generated code.
What Is Multi-Agent Coding? Architecture, Workflows, and Examples with Codex, Claude Code, and GitHub CopilotA practical guide to multi-agent coding architecture, workflows, tradeoffs, and implementation patterns in Codex, Claude Code, and GitHub Copilot.
Plan Mode vs. Autopilot: Codex, Claude Code, and CopilotStart Implementation or Start with Autopilot in VS Code? Compare these Copilot choices with Codex Plan Mode and Claude Code, and learn when a plan needs review.
Does Codex Have Remote Control? Codex vs. Claude Code vs. GitHub CopilotYes. Codex remote access connects a ChatGPT desktop host to mobile or another supported desktop. Compare it with Claude Code and Copilot, including VS Code.
Instructions vs. Skills: What Belongs Where?A practical guide to choosing between project instructions, reusable agent skills, one-off prompts, and deterministic tooling, with examples for Codex and Claude Code.
Native UI or CLI? Choosing the right interface for Claude Code and CodexA practical comparison of desktop and terminal workflows in Claude Code and Codex, including their trade-offs, best use cases, and a hybrid setup that works.
AGENTS.md across Codex, Claude Code, and GitHub Copilot: what belongs in the fileHow Codex, Claude Code, and GitHub Copilot load repository instructions, what the Karpathy-inspired rules actually are, and how to write guidance that helps instead of bloating every task.
Your coding agent’s output is part of the interfaceHow action-first, low-friction responses can make a coding agent easier to use when long explanations get in the way of the next useful step.
The agent harness is the new framework: how to make coding agents reliableA practical guide to the planning, isolation, state, evaluation, and permission boundaries that turn a capable coding model into a reliable engineering system.
Loop engineering in Claude Code: keep documentation aligned with the codebaseA practical documentation drift loop for Claude Code: detect changes every two hours, review the full picture weekly, and keep humans in control.