https://dopp.sh/blog/dopp sits between your app and Jev, or any API that answers the same way. You change one URL, your app gets the same answers, and every request is kept. You train a small model of your own on those requests, with labels from you, an LLM, a dataset or another API, and run it hosted, on your own servers, or offline in the browser, with your API as the fallback.en https://dopp.sh/blog/fine-tune-kev/https://dopp.sh/blog/fine-tune-kev/Fri, 02 Oct 2026 12:00:00 GMTKev 9B went from 57.5% to 75% on Kev's own emotion eval after training on 2,500 labelled rows. How, start to finish, for about $1 of GPU. https://dopp.sh/blog/gatekeeper/https://dopp.sh/blog/gatekeeper/Tue, 29 Sep 2026 12:00:00 GMTI prototyped a castle guard on Jev, trained a 34 MB model on a few hundred labelled lines, and now the guard runs in your browser tab with no server. https://dopp.sh/blog/run-jev-offline/https://dopp.sh/blog/run-jev-offline/Tue, 29 Sep 2026 12:00:00 GMTTrain a 34 MB model on the requests your app already sends Jev and run it with no internet. I did it with Home Assistant for $0.12. https://dopp.sh/blog/stop-paying-for-decisions/https://dopp.sh/blog/stop-paying-for-decisions/Thu, 24 Sep 2026 12:00:00 GMTChange one URL, keep the decision API you use today, and by the afternoon own a small model trained on your own requests. https://dopp.sh/blog/every-request-is-training-data/https://dopp.sh/blog/every-request-is-training-data/Thu, 24 Sep 2026 12:00:00 GMTA fine-tuned model needs examples. You already produce hundreds a day. A proxy keeps every one, in your format, and you pick what labels them. https://dopp.sh/blog/vs-huggingface-autotrain/https://dopp.sh/blog/vs-huggingface-autotrain/Thu, 24 Sep 2026 12:00:00 GMTAutoTrain fine-tunes a Hub model on a CSV you upload. dopp makes the CSV from your live traffic, labels it, trains a model, measures it, and routes. https://dopp.sh/blog/vs-jev/https://dopp.sh/blog/vs-jev/Thu, 24 Sep 2026 12:00:00 GMTJev answers typed questions with a hosted model. dopp keeps Jev as the fallback and turns your Jev traffic into a small model you own, hosted or offline. https://dopp.sh/blog/vs-openai-fine-tuning/https://dopp.sh/blog/vs-openai-fine-tuning/Thu, 24 Sep 2026 12:00:00 GMTOpenAI fine-tunes their models on your JSONL. dopp collects the JSONL from your live traffic, labels it, trains an open model you own, and measures it. https://dopp.sh/blog/vs-simple-jev-rfdt/https://dopp.sh/blog/vs-simple-jev-rfdt/Thu, 24 Sep 2026 12:00:00 GMTsimple-jev answers typed questions with open models; RFDT is its training script. dopp is the product around that idea: capture, labels, training, routing.