Writekin
fine-tune a local LLM on your own writing, on your Mac
Details
- External ID
- 49088436
- Source
- HN
- Company
- —
- Product
- Writekin
- Website domain
- github.com
- Launched
- July 28, 2026
- Cohort
- —
- Upvotes
- 6
- Upvotes percentile
- 0.2873357228195938
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
Hey Hacker News!I 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.The usual fixes I found online for this were:- Some sort of SKILL.md, or- 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).While 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.So 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.Everything runs on your Mac. Ingestion, training, and generation are all local. The only network calls are:(1) When you download the model weights from Hugging Face and(2) The Sparkle update check.Training 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!Quick 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.Would genuinely love feedback!Source: https://github.com/scouttyg/writekin
Enrichment
- Theme
- ai voice dictation and transcription tools
- Vertical
- Horizontal
- Function
- Content generation
- Audience
- Prosumer
- AI stance
- AI feature
- Project type
- Commercial product
- Normalized one-liner
- fine-tune local llm on your writing
- Manually corrected
- False
Could you build this?
Yes A Mac wrapper executing local LoRA or QLoRA fine-tuning scripts via MLX or PyTorch and serving the result via llama.cpp is standard and easily scaffolded.
Discussion
3 comments analyzed.
Concerns raised: High hardware requirements for inference on lower-spec servers, Model size and RAM requirements unclear for different deployment scenarios
Feature requests: Support for model export/transfer to cheaper cloud servers for inference, Clarification on inference requirements vs training requirements
Competitors
Other products that read as similar to this one — 135 launches clear the similarity bar, closest 8 shown.
Attention rank: #109 of 136 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 263 days after the earliest competitor.
- Write in your unique voice, with AI critique · hn · 2026-05-06 · 5 upvotes · similarity 0.51
- TongueType · hn · 2026-05-15 · 5 upvotes · similarity 0.46
- CyberWriter · hn · 2026-04-20 · 15 upvotes · similarity 0.45
- Gemma 4 Multimodal Fine-Tuner for Apple Silicon · hn · 2026-04-07 · 235 upvotes · similarity 0.44
- threadfork · hn · 2026-07-21 · 10 upvotes · similarity 0.44
- Reproof · hn · 2026-08-18 · 5 upvotes · similarity 0.41
- Droptheslop.ai · hn · 2026-02-16 · 6 upvotes · similarity 0.41
- I built a free CharacterAI that runs locally · hn · 2026-03-25 · 8 upvotes · similarity 0.40
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a content generation tool for Government yet.