<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Ahmet Canik</title><description>Thoughts on AI and its impact, mostly on software engineering, but not exclusively.</description><link>https://example.com/</link><item><title>The Limits of Vibe Coding: Why AI Still Doesn&apos;t Replace Senior Engineers</title><link>https://example.com/blog/01-the-limits-of-vibe-coding/</link><guid isPermaLink="true">https://example.com/blog/01-the-limits-of-vibe-coding/</guid><description>AI walks the decision tree. Weighting it, knowing which branches cost you later, is still yours. That is where senior engineering judgment lives.</description><pubDate>Tue, 07 Apr 2026 00:00:00 GMT</pubDate></item><item><title>The 20-Minute Repricing</title><link>https://example.com/blog/02-the-20-minute-repricing/</link><guid isPermaLink="true">https://example.com/blog/02-the-20-minute-repricing/</guid><description>A manager builds a working prototype in 20 minutes with an AI agent and quietly reprices engineering. The hidden costs are real. The bill is still coming.</description><pubDate>Thu, 09 Apr 2026 00:00:00 GMT</pubDate></item><item><title>The Last Post-Mortem</title><link>https://example.com/blog/03-the-last-post-mortem/</link><guid isPermaLink="true">https://example.com/blog/03-the-last-post-mortem/</guid><description>AI became competent because developers documented every failure at 2am. Now the agent answers before the post gets written. The feedback loop is quietly dying.</description><pubDate>Wed, 15 Apr 2026 00:00:00 GMT</pubDate></item><item><title>The Edge Case</title><link>https://example.com/blog/04-the-edge-case/</link><guid isPermaLink="true">https://example.com/blog/04-the-edge-case/</guid><description>Software does not fail on the average case. AI models are losing their grip on the tail distribution, exactly as humans stop writing about what breaks them.</description><pubDate>Mon, 20 Apr 2026 00:00:00 GMT</pubDate></item></channel></rss>