{"slug": "show-hn-writekin-fine-tune-a-local-llm-on-your-own-writing-on-your-mac", "title": "Show HN: Writekin – fine-tune a local LLM on your own writing, on your Mac", "summary": "Writekin, a new open-source macOS app, fine-tunes a local LLM on a user's own writing from Apple Mail, iMessage, and local documents to generate text in their personal voice, with all processing running on-device via Apple's MLX. Developer Scott G. released the v0.9 prototype on GitHub, noting that output quality is uneven but demonstrates the feasibility of personalized local AI writing assistance.", "body_md": "Hey Hacker News!\n\nI built Writekin over the past week because I was tired of AI writing that didn't sound like me, even though I had just used AI to clean it up, rather than wholesale write it.\n\nThe usual fixes I found online for this were:\n\n- Some sort of SKILL.md, or\n\n- A system prompt full of rules to strip the generic AI tells (e.g. no em-dashes, none of the stock phrases, varying the sentence length, etc).\n\nWhile those cleaned up the surface a bit, Pangram still came back as ~100% AI written, which was frustrating, as again it was mainly taking my sloppy copy and tweaking it.\n\nSo when building Writekin I took a different route: Writekin fine-tunes a local model on your own writing. It reads what you've already written (Apple Mail, iMessage, local documents, chat exports), curates it into a training corpus, and runs a QLoRA fine-tuning on-device via Apple's MLX. A Compose screen then drafts and rewrites in that voice.\n\nEverything runs on your Mac. Ingestion, training, and generation are all local. The only network calls are:\n\n(1) When you download the model weights from Hugging Face and\n\n(2) The Sparkle update check.\n\nTraining on your own Mail/Messages only felt okay to ship if the end user could verify that, so the source is public — so you can read exactly what it does!\n\nQuick gut check: It's v0.9 and the output is uneven. Honestly, sometimes it nails your voice, and sometimes it's just completely off. This is more a \"this is possible and kind of works\" than a finished product.\n\nWould genuinely love feedback!\n\nSource: [https://github.com/scouttyg/writekin](https://github.com/scouttyg/writekin)\n\nComments URL: [https://news.ycombinator.com/item?id=49088436](https://news.ycombinator.com/item?id=49088436)\n\nPoints: 1\n\n# Comments: 0", "url": "https://wpnews.pro/news/show-hn-writekin-fine-tune-a-local-llm-on-your-own-writing-on-your-mac", "canonical_source": "https://github.com/scouttyg/writekin", "published_at": "2026-07-28 19:05:12+00:00", "updated_at": "2026-07-28 19:22:37.001172+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-tools", "ai-products", "developer-tools"], "entities": ["Writekin", "Apple", "Hugging Face", "MLX", "GitHub", "Scott G."], "alternates": {"html": "https://wpnews.pro/news/show-hn-writekin-fine-tune-a-local-llm-on-your-own-writing-on-your-mac", "markdown": "https://wpnews.pro/news/show-hn-writekin-fine-tune-a-local-llm-on-your-own-writing-on-your-mac.md", "text": "https://wpnews.pro/news/show-hn-writekin-fine-tune-a-local-llm-on-your-own-writing-on-your-mac.txt", "jsonld": "https://wpnews.pro/news/show-hn-writekin-fine-tune-a-local-llm-on-your-own-writing-on-your-mac.jsonld"}}