<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Tudor Turcanu — Blog</title><description>Writing about iOS development, Swift, SwiftUI, and on-device AI.</description><link>https://tudorturcanu.ch/</link><item><title>Streaming LLM Responses with Swift Concurrency</title><link>https://tudorturcanu.ch/blog/streaming-llm-swift-concurrency/</link><guid isPermaLink="true">https://tudorturcanu.ch/blog/streaming-llm-swift-concurrency/</guid><description>Token streams are a natural fit for AsyncSequence — until cancellation, actor isolation, and UI update frequency get involved. Here&apos;s the shape that works.</description><pubDate>Wed, 08 Jul 2026 00:00:00 GMT</pubDate></item></channel></rss>