<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Zidane — Tech &amp; AI Blog</title><link>https://zidanewu.com/</link><description>Recent content on Zidane — Tech &amp; AI Blog</description><image><title>Zidane — Tech &amp; AI Blog</title><url>https://zidanewu.com/og-default.png</url><link>https://zidanewu.com/og-default.png</link></image><generator>Hugo -- 0.146.0</generator><language>en-us</language><lastBuildDate>Sun, 29 Mar 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://zidanewu.com/index.xml" rel="self" type="application/rss+xml"/><item><title>A Hidden Comment in a GitHub Issue Almost Owned Our CI Pipeline</title><link>https://zidanewu.com/posts/ai-coding-agent-prompt-injection-security/</link><pubDate>Sun, 29 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-coding-agent-prompt-injection-security/</guid><description>Prompt injection attacks on AI coding agents are real and growing. Here&amp;#39;s how a hidden payload in a GitHub issue nearly compromised our build system.</description></item><item><title>My Best Engineer Just Told Me AI Made His Job Boring</title><link>https://zidanewu.com/posts/ai-agents-taking-fun-out-of-programming/</link><pubDate>Sun, 29 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-agents-taking-fun-out-of-programming/</guid><description>AI coding agents handle the creative parts of engineering while humans review diffs. A tech lead&amp;#39;s honest account of managing team morale when the fun parts disappear.</description></item><item><title>I Replaced Our Figma-to-Dev Handoff With Google Stitch. Here's What Broke</title><link>https://zidanewu.com/posts/google-stitch-review-tech-lead/</link><pubDate>Sun, 22 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/google-stitch-review-tech-lead/</guid><description>A tech lead&amp;#39;s honest review of Google Stitch 2.0&amp;#39;s AI-native canvas after testing it on three real projects. What works, what doesn&amp;#39;t, and who should actually use it.</description></item><item><title>My Team Tried 'Agentic Engineering' for Two Sprints. It's Not What You Think</title><link>https://zidanewu.com/posts/agentic-engineering-vs-vibe-coding/</link><pubDate>Sun, 22 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/agentic-engineering-vs-vibe-coding/</guid><description>Agentic engineering is the hot new term for professional AI-assisted coding. After running it on real projects, here&amp;#39;s what the evangelists aren&amp;#39;t telling you.</description></item><item><title>AI Code Review vs Human Code Review: What I Learned From 100 PRs</title><link>https://zidanewu.com/posts/ai-code-review-vs-human-100-prs/</link><pubDate>Fri, 20 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-code-review-vs-human-100-prs/</guid><description>I ran an informal A/B test across 100 PRs, routing each through both AI and human review. AI caught 23 bugs humans missed. It also produced hundreds of useless comments. Here&amp;rsquo;s what the data actually says.</description></item><item><title>Building an AI Coding Standard for Your Engineering Team</title><link>https://zidanewu.com/posts/ai-coding-standards-team-guide/</link><pubDate>Fri, 20 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-coding-standards-team-guide/</guid><description>Ad-hoc AI usage creates inconsistency and hidden risk. Here&amp;#39;s how to build a practical standard: AGENTS.md, .cursorrules, prompt libraries, and review checklists — with real templates.</description></item><item><title>How I Use AI to Prep for 1:1s (Without Being Creepy About It)</title><link>https://zidanewu.com/posts/ai-1on1-meeting-prep/</link><pubDate>Fri, 20 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-1on1-meeting-prep/</guid><description>AI can help you prep for 1:1s by surfacing action items, recent PRs, and discussion topics. But there&amp;rsquo;s a line you shouldn&amp;rsquo;t cross: never let AI write your actual feedback.</description></item><item><title>I Automated Our Standup Notes With AI. My Team Almost Mutinied</title><link>https://zidanewu.com/posts/ai-standup-automation-gone-wrong/</link><pubDate>Fri, 20 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-standup-automation-gone-wrong/</guid><description>Automating standup notes with AI seemed like a small quality-of-life win. It turned out to surface real questions about surveillance, trust, and what standup is actually for.</description></item><item><title>I Let AI Write Our Technical RFC. Here's What Happened</title><link>https://zidanewu.com/posts/ai-rfc-writing-experiment/</link><pubDate>Fri, 20 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-rfc-writing-experiment/</guid><description>Claude drafted a technical RFC in 20 minutes. My teammates said it &amp;#39;reads like a textbook.&amp;#39; Here&amp;#39;s what the experiment taught us about where AI fits in technical writing.</description></item><item><title>MCP Servers I Actually Use Daily (And 5 That Were Overhyped)</title><link>https://zidanewu.com/posts/mcp-servers-daily-use/</link><pubDate>Fri, 20 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/mcp-servers-daily-use/</guid><description>After running MCP servers in my daily workflow for months, here&amp;#39;s what&amp;#39;s genuinely useful versus what sounded great in a demo and quietly disappeared from my setup.</description></item><item><title>The Hidden Cost of AI-Generated Tests — Why 90% Coverage Means Nothing</title><link>https://zidanewu.com/posts/ai-generated-tests-hidden-cost/</link><pubDate>Fri, 20 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-generated-tests-hidden-cost/</guid><description>AI can write tests in seconds, but coverage numbers are lying to you. Here&amp;#39;s how tautological, implementation-coupled tests let a real bug survive for three weeks.</description></item><item><title>The Practical Guide to Using AI on Legacy Codebases (Without Making Things Worse)</title><link>https://zidanewu.com/posts/ai-legacy-code-migration-guide/</link><pubDate>Fri, 20 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-legacy-code-migration-guide/</guid><description>AI tools can genuinely help with legacy code — but only on the right problems. Understanding and documentation: yes. Refactoring core logic you don&amp;rsquo;t fully understand: hard no.</description></item><item><title>Why I Stopped Letting AI Write Database Migrations</title><link>https://zidanewu.com/posts/ai-database-migration-cautionary-tale/</link><pubDate>Fri, 20 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-database-migration-cautionary-tale/</guid><description>An AI-generated migration ran perfectly in dev. In production, it would have locked a large production table during business hours. What I learned the hard way about AI and database changes.</description></item><item><title>Why Senior Engineers Are the Biggest AI Skeptics (And Why They're Half Right)</title><link>https://zidanewu.com/posts/senior-engineers-ai-skeptics/</link><pubDate>Fri, 20 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/senior-engineers-ai-skeptics/</guid><description>Senior engineers have seen enough hype cycles to be skeptical of AI tools — and their concerns about maintainability and debugging complexity are legitimate. But refusing to engage is its own kind of risk.</description></item><item><title>I Ran Codex and Claude Code Side by Side for a Week</title><link>https://zidanewu.com/posts/codex-vs-claude-code-2026/</link><pubDate>Thu, 19 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/codex-vs-claude-code-2026/</guid><description>Codex just hit 1M weekly active users. Claude Code has been my daily driver for months. I gave both the same tasks for a week and tracked everything.</description></item><item><title>The NYT Says Coding Is Dead. Here's What They Got Wrong (And Right)</title><link>https://zidanewu.com/posts/nyt-coding-dead-response/</link><pubDate>Thu, 19 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/nyt-coding-dead-response/</guid><description>The NYT&amp;#39;s &amp;#39;Coding After Coders&amp;#39; piece made a lot of engineers defensive. As a Tech Lead managing a small engineering team, I read it twice. Here&amp;#39;s my honest take.</description></item><item><title>Every AI Coding Tool Switched to Credits. My Team's Bill Tripled.</title><link>https://zidanewu.com/posts/ai-coding-tool-credit-pricing-trap/</link><pubDate>Sun, 15 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-coding-tool-credit-pricing-trap/</guid><description>How credit-based pricing took over AI coding tools and why your team is probably paying 3x more than the sticker price.</description></item><item><title>I Tracked My Team's AI Tool Switching for 3 Months. The Productivity Loss Was Staggering.</title><link>https://zidanewu.com/posts/ai-tool-switching-cost-teams/</link><pubDate>Sun, 15 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-tool-switching-cost-teams/</guid><description>The hidden cost of AI tool churn: context switching, config drift, and the productivity tax nobody talks about.</description></item><item><title>Cursor Automations: I Let Always-On Agents Run My Code Reviews for a Week</title><link>https://zidanewu.com/posts/cursor-automations-always-on-agents/</link><pubDate>Sun, 08 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/cursor-automations-always-on-agents/</guid><description>A practical look at Cursor&amp;#39;s new Automations feature — always-on coding agents triggered by PRs, Slack, and PagerDuty. What worked, what broke, and what it means for engineering teams.</description></item><item><title>GPT-5.4 for Coding: My First 72 Hours With OpenAI's New Model</title><link>https://zidanewu.com/posts/gpt-5-4-coding-first-look/</link><pubDate>Sun, 08 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/gpt-5-4-coding-first-look/</guid><description>A hands-on first look at GPT-5.4&amp;#39;s coding capabilities — benchmarks, real-world impressions, and where it still falls short compared to Claude Opus 4.6.</description></item><item><title>I Added an AI to Every PR for 30 Days. Here's What Actually Caught Bugs</title><link>https://zidanewu.com/posts/ai-pr-review-bots-comparison/</link><pubDate>Sun, 01 Mar 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-pr-review-bots-comparison/</guid><description>A hands-on comparison of CodeRabbit, Qodo, Ellipsis, and BugBot after 30 days of running them on real pull requests. Real bug counts, false positives, and honest verdicts.</description></item><item><title>My Team Vibe-Coded a Feature. Three Weeks Later, I'm Still Cleaning Up</title><link>https://zidanewu.com/posts/vibe-coding-technical-debt/</link><pubDate>Sat, 28 Feb 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/vibe-coding-technical-debt/</guid><description>What happens when you let AI generate a feature end-to-end without governance. A tech lead&amp;#39;s honest account of the fallout.</description></item><item><title>Block Cut ~4,000 Roles and Cited AI Efficiency — What Leaders Should Do Next</title><link>https://zidanewu.com/posts/block-layoffs-ai-white-collar-panic/</link><pubDate>Fri, 27 Feb 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/block-layoffs-ai-white-collar-panic/</guid><description>Block cut 4,000 employees citing AI efficiency. What does this mean for white-collar workers, and how do you stay relevant?</description></item><item><title>Claude Code Review Checklist (2026): 12 Checks Before You Approve</title><link>https://zidanewu.com/posts/claude-code-review-checklist-2026/</link><pubDate>Tue, 24 Feb 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/claude-code-review-checklist-2026/</guid><description>A practical Claude Code review checklist for tech leads: 12 checks that catch real bugs before merge.</description></item><item><title>I Ran 5 AI Code Review Tools on the Same PR. Only 2 Were Useful</title><link>https://zidanewu.com/posts/ai-review-agents-same-pr-test/</link><pubDate>Mon, 23 Feb 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-review-agents-same-pr-test/</guid><description>I tested five AI code review agents on one messy PR and compared signal, noise, and review time savings.</description></item><item><title>30 Fake AI Extensions Were Stealing My Team's Data — How We Caught It</title><link>https://zidanewu.com/posts/fake-ai-chrome-extensions-developer-security/</link><pubDate>Fri, 20 Feb 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/fake-ai-chrome-extensions-developer-security/</guid><description>Fake AI Chrome extensions are stealing developer API keys and credentials. Here&amp;#39;s how my team found one, what we learned, and how to audit yours.</description></item><item><title>I Tried Figma's Code to Canvas With My Team — Here's What Actually Changed</title><link>https://zidanewu.com/posts/figma-code-to-canvas-review/</link><pubDate>Wed, 18 Feb 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/figma-code-to-canvas-review/</guid><description>Hands-on review of Figma&amp;#39;s new Code to Canvas feature powered by Anthropic. What works, what breaks, and whether it actually fixes the design-dev gap.</description></item><item><title>I Let GitHub's Agentic Workflows Run My Repo for a Week. Here's What Happened</title><link>https://zidanewu.com/posts/github-agentic-workflows-review/</link><pubDate>Sun, 15 Feb 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/github-agentic-workflows-review/</guid><description>A hands-on review of GitHub Agentic Workflows after running them on a real repo for a week — the wins, the surprises, and the stuff that&amp;#39;s not ready yet.</description></item><item><title>5 Ways I Use AI as an Engineering Manager (That Aren't Coding)</title><link>https://zidanewu.com/posts/engineering-manager-ai-tools-2026/</link><pubDate>Sat, 14 Feb 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/engineering-manager-ai-tools-2026/</guid><description>AI isn&amp;#39;t just for writing code. Here&amp;#39;s how I use it for 1:1s, planning, documentation, and the other 60% of a tech lead&amp;#39;s job.</description></item><item><title>I Spent 3 Hours Configuring MCP Servers Across 4 Editors. Then I Found add-mcp</title><link>https://zidanewu.com/posts/mcp-fragmentation-add-mcp/</link><pubDate>Fri, 13 Feb 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/mcp-fragmentation-add-mcp/</guid><description>I wasted 3 hours configuring MCP servers across 4 editors. Then I found add-mcp, a CLI that does it in one command.</description></item><item><title>Prompt Engineering for Developers: 7 Patterns That Actually Work</title><link>https://zidanewu.com/posts/prompt-engineering-developers-2026/</link><pubDate>Tue, 10 Feb 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/prompt-engineering-developers-2026/</guid><description>Skip the prompt engineering hype. Here are 7 concrete patterns that make AI coding tools significantly more useful.</description></item><item><title>How I Use AI to Review Code 3x Faster (Without Missing Bugs)</title><link>https://zidanewu.com/posts/ai-code-review-2026/</link><pubDate>Tue, 03 Feb 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-code-review-2026/</guid><description>A practical guide to using AI for code review — what actually works, what doesn&amp;#39;t, and where human judgment still matters.</description></item><item><title>What AI Gets Wrong About Code Generation (And How to Fix It)</title><link>https://zidanewu.com/posts/ai-code-generation-mistakes/</link><pubDate>Wed, 28 Jan 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-code-generation-mistakes/</guid><description>AI-generated code has predictable failure modes. Here are the 8 most common mistakes I see and the workarounds I&amp;#39;ve developed.</description></item><item><title>Cursor vs Claude Code in 2026: Which AI Coding Tool Should You Use?</title><link>https://zidanewu.com/posts/cursor-vs-claude-code-2026/</link><pubDate>Sun, 25 Jan 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/cursor-vs-claude-code-2026/</guid><description>An honest comparison of Cursor and Claude Code from someone who uses both daily. When to use which, and why you probably need both.</description></item><item><title>Designing Better APIs with AI: Patterns That Work</title><link>https://zidanewu.com/posts/ai-api-design/</link><pubDate>Thu, 22 Jan 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-api-design/</guid><description>How I use AI to design, review, and document APIs — the patterns that save time and the ones that waste it.</description></item><item><title>Building an AI-Powered Development Workflow from Scratch</title><link>https://zidanewu.com/posts/ai-development-workflow/</link><pubDate>Tue, 20 Jan 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-development-workflow/</guid><description>A complete walkthrough of integrating AI into every phase of software development — from planning to deploy. What works, what doesn&amp;#39;t, and the workflow I&amp;#39;ve settled on.</description></item><item><title>10 AI Coding Assistant Tips That Actually Save Me Hours Every Week</title><link>https://zidanewu.com/posts/claude-code-tips-2026/</link><pubDate>Sat, 17 Jan 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/claude-code-tips-2026/</guid><description>Practical AI coding assistant tips from a Tech Lead who uses them daily. No fluff — just real workflows that work with Claude Code, Copilot Chat, Cursor, and similar tools.</description></item><item><title>AI-Assisted Incident Response: A Practical Playbook</title><link>https://zidanewu.com/posts/ai-incident-response/</link><pubDate>Thu, 15 Jan 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-incident-response/</guid><description>How we integrated AI into our incident response process — what accelerated resolution and what made things worse.</description></item><item><title>AI Tools That Actually Help Remote Engineering Teams</title><link>https://zidanewu.com/posts/ai-remote-engineering-teams/</link><pubDate>Mon, 12 Jan 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-remote-engineering-teams/</guid><description>Not another list of productivity apps. These are the AI workflows that genuinely improve async communication, code review, and sprint planning for distributed teams.</description></item><item><title>The AI Tools I Actually Use as a Tech Lead (2026 Edition)</title><link>https://zidanewu.com/posts/ai-tools-tech-lead-2026/</link><pubDate>Thu, 08 Jan 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-tools-tech-lead-2026/</guid><description>Not a listicle of 50 tools you&amp;#39;ll never try. These are the 7 AI tools I use daily as a Tech Lead managing engineers and data scientists.</description></item><item><title>Using AI to Onboard New Engineers 2x Faster</title><link>https://zidanewu.com/posts/ai-onboarding-engineers/</link><pubDate>Tue, 06 Jan 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-onboarding-engineers/</guid><description>We cut our engineering onboarding time from 6-8 weeks to about 3 weeks. AI wasn&amp;#39;t the only factor, but it was the biggest one.</description></item><item><title>How I Write Technical Documents 3x Faster with AI</title><link>https://zidanewu.com/posts/writing-technical-docs-with-ai/</link><pubDate>Sat, 03 Jan 2026 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/writing-technical-docs-with-ai/</guid><description>Practical templates and prompts for writing RFCs, ADRs, runbooks, and other technical documents with AI assistance.</description></item><item><title>Local LLMs vs Cloud APIs: When to Use Which for Development</title><link>https://zidanewu.com/posts/local-llms-vs-cloud/</link><pubDate>Sun, 28 Dec 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/local-llms-vs-cloud/</guid><description>I run local LLMs for some tasks and cloud APIs for others. Here&amp;#39;s my framework for deciding which to use when.</description></item><item><title>AI Pair Programming in 2026: A Practical Guide</title><link>https://zidanewu.com/posts/ai-pair-programming-guide/</link><pubDate>Mon, 22 Dec 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-pair-programming-guide/</guid><description>How to use AI as a genuine pair programming partner — workflow design, prompting strategies, and when to ditch the AI and think for yourself.</description></item><item><title>GitHub Copilot in 2026: Is It Still Worth $19/Month?</title><link>https://zidanewu.com/posts/github-copilot-review-2026/</link><pubDate>Sat, 20 Dec 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/github-copilot-review-2026/</guid><description>GitHub Copilot was the first AI coding tool most of us used. With Cursor, Claude Code, and others now available, is it still worth paying for?</description></item><item><title>How AI Transformed Our Sprint Planning (Lessons from 6 Months)</title><link>https://zidanewu.com/posts/ai-sprint-planning/</link><pubDate>Thu, 18 Dec 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-sprint-planning/</guid><description>We&amp;#39;ve been using AI to assist with sprint planning for six months. Some things worked surprisingly well. Others were a waste of time.</description></item><item><title>Windsurf vs Cursor in 2026: Which AI IDE Should You Pick?</title><link>https://zidanewu.com/posts/windsurf-vs-cursor-2026/</link><pubDate>Mon, 15 Dec 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/windsurf-vs-cursor-2026/</guid><description>I used Windsurf and Cursor side by side for 30 days. Here&amp;#39;s which AI-powered IDE is better for different types of developers.</description></item><item><title>How I Use AI to Debug Production Issues (Real Examples)</title><link>https://zidanewu.com/posts/ai-debugging-production/</link><pubDate>Fri, 12 Dec 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-debugging-production/</guid><description>Real production debugging stories using AI tools — stack traces, log analysis, and the mistakes I made along the way.</description></item><item><title>Using AI to Refactor Legacy Code Without Breaking Everything</title><link>https://zidanewu.com/posts/ai-refactoring-legacy-code/</link><pubDate>Mon, 08 Dec 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-refactoring-legacy-code/</guid><description>Legacy code is scary. AI can help you understand it, test it, and refactor it — if you&amp;#39;re careful about how you use it.</description></item><item><title>5 Free AI Coding Tools That Are Actually Worth Using in 2026</title><link>https://zidanewu.com/posts/free-ai-coding-tools-2026/</link><pubDate>Fri, 05 Dec 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/free-ai-coding-tools-2026/</guid><description>Not every AI coding tool needs a $20/month subscription. These 5 free tools are genuinely useful for developers on a budget.</description></item><item><title>GitHub Copilot vs Cursor vs Claude Code: Which AI Coding Tool Wins?</title><link>https://zidanewu.com/posts/copilot-vs-cursor-vs-claude-code/</link><pubDate>Mon, 01 Dec 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/copilot-vs-cursor-vs-claude-code/</guid><description>A hands-on comparison of the three biggest AI coding tools. Real code scenarios, real opinions, real costs.</description></item><item><title>Understanding LLM Context Windows: A Developer's Practical Guide</title><link>https://zidanewu.com/posts/llm-context-window-guide/</link><pubDate>Fri, 28 Nov 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/llm-context-window-guide/</guid><description>Context windows are the single most important LLM concept for developers to understand. Here&amp;#39;s what I wish someone had told me earlier.</description></item><item><title>How AI Changed My Testing Strategy (For the Better)</title><link>https://zidanewu.com/posts/ai-testing-strategies/</link><pubDate>Thu, 20 Nov 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/ai-testing-strategies/</guid><description>After 6 months of integrating AI into our testing workflow, here&amp;#39;s what actually improved — and what surprised me.</description></item><item><title>ChatGPT vs Claude in 2026: An Honest Comparison for Developers</title><link>https://zidanewu.com/posts/chatgpt-vs-claude-2026/</link><pubDate>Sat, 15 Nov 2025 00:00:00 +0000</pubDate><guid>https://zidanewu.com/posts/chatgpt-vs-claude-2026/</guid><description>A real-world comparison of ChatGPT and Claude for software development — coding, reasoning, context handling, and pricing. No fanboys, just honest experience.</description></item><item><title>About</title><link>https://zidanewu.com/about/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://zidanewu.com/about/</guid><description>&lt;h2 id="about-this-blog">About This Blog&lt;/h2>
&lt;p>I&amp;rsquo;m an engineering lead working on data infrastructure at a mid-size tech company. I started this blog because I kept having the same conversations with other tech leads about which AI tools are actually useful versus which ones just demo well.&lt;/p>
&lt;p>Most of what I write comes from my day-to-day: managing a team, reviewing code, writing RFCs, and trying to get more done without burning out. I test a lot of tools and only write about the ones that stick around in my workflow for more than a week.&lt;/p></description></item><item><title>AI Code Review Prompt Pack (Free)</title><link>https://zidanewu.com/resources/code-review-prompts/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://zidanewu.com/resources/code-review-prompts/</guid><description>10 copy-paste prompts for AI-assisted code review. Battle-tested on 500+ PRs.</description></item></channel></rss>