DEdebo_·May 12, 2026Submission
Kiln – a free app and open-source library to build better AI products
2 points · Open discussionKiln-AI
Kiln
Python2024–2026
Kiln-AI / Kiln
Kiln
Build, Evaluate, and Optimize AI Systems. Includes evals, RAG, agents, fine-tuning, synthetic data generation, dataset management, MCP, and more.
4Distinct people
15Mentions
1Explicit recommenders
5.3KGitHub stars
What to know before exploring
What it is
Build, Evaluate, and Optimize AI Systems. Includes evals, RAG, agents, fine-tuning, synthetic data generation, dataset management, MCP, and more.
GitHub
Where it is used
No documented product use has been verified yet.
What people are saying
Fine tuning to a specific codebase is a bit strange. It's going to learn some style/tool guidance which is good (but there are other ways of getting), at the risk of unlearning…
scosman used a recommendation phrase. One indexed comment matches. This is a phrase match, not a verdict.
No classified warning is attached to this page. Read the discussions before treating silence as approval.
Also in these discussions
No other repository is repeatedly linked in the same discussions.
This week on Hacker News
Nov 7–Nov 14 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/14/2026, 5:04:57 PM UTC.
No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.
Mentions over time
HN mentions by month15 dated mentions
Observed mentions: 2024-11-21 15:47:43 UTC to 2026-05-12 17:59:27 UTC.
Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.
View mention counts and date ranges
| Period | Start (UTC) | End (exclusive, UTC) | Mentions | Submissions | Comments |
|---|---|---|---|---|---|
| Nov 2024 | 2024-11-01 00:00:00 UTC | 2024-12-01 00:00:00 UTC | 1 | 1 | 0 |
| Dec 2024 | 2024-12-01 00:00:00 UTC | 2025-01-01 00:00:00 UTC | 1 | 1 | 0 |
| Jan 2025 | 2025-01-01 00:00:00 UTC | 2025-02-01 00:00:00 UTC | 1 | 0 | 1 |
| Feb 2025 | 2025-02-01 00:00:00 UTC | 2025-03-01 00:00:00 UTC | 1 | 1 | 0 |
| Mar 2025 | 2025-03-01 00:00:00 UTC | 2025-04-01 00:00:00 UTC | 3 | 1 | 2 |
| Apr 2025 | 2025-04-01 00:00:00 UTC | 2025-05-01 00:00:00 UTC | 1 | 0 | 1 |
| May 2025 | 2025-05-01 00:00:00 UTC | 2025-06-01 00:00:00 UTC | 0 | 0 | 0 |
| Jun 2025 | 2025-06-01 00:00:00 UTC | 2025-07-01 00:00:00 UTC | 2 | 0 | 2 |
| Jul 2025 | 2025-07-01 00:00:00 UTC | 2025-08-01 00:00:00 UTC | 1 | 1 | 0 |
| Aug 2025 | 2025-08-01 00:00:00 UTC | 2025-09-01 00:00:00 UTC | 0 | 0 | 0 |
| Sep 2025 | 2025-09-01 00:00:00 UTC | 2025-10-01 00:00:00 UTC | 1 | 0 | 1 |
| Oct 2025 | 2025-10-01 00:00:00 UTC | 2025-11-01 00:00:00 UTC | 0 | 0 | 0 |
| Nov 2025 | 2025-11-01 00:00:00 UTC | 2025-12-01 00:00:00 UTC | 0 | 0 | 0 |
| Dec 2025 | 2025-12-01 00:00:00 UTC | 2026-01-01 00:00:00 UTC | 0 | 0 | 0 |
| Jan 2026 | 2026-01-01 00:00:00 UTC | 2026-02-01 00:00:00 UTC | 1 | 0 | 1 |
| Feb 2026 | 2026-02-01 00:00:00 UTC | 2026-03-01 00:00:00 UTC | 1 | 0 | 1 |
| Mar 2026 | 2026-03-01 00:00:00 UTC | 2026-04-01 00:00:00 UTC | 0 | 0 | 0 |
| Apr 2026 | 2026-04-01 00:00:00 UTC | 2026-05-01 00:00:00 UTC | 0 | 0 | 0 |
| May 2026 | 2026-05-01 00:00:00 UTC | 2026-06-01 00:00:00 UTC | 1 | 1 | 0 |
What Hacker News says
15 sourcesI make a project for evals and fine-tuning and our default example task is a joke generator. It's a fun demo, but more importantly it's a really good use case to show how evaluating and optimizing LLMs is hard.
- There are a dozen plus common failure modes. How you split setup/punchline. Tropes. Toxicity. Template reuse. Each one needs a good eval.
- Datasets are hard: there's not much off the shelf, and as this author points out scraping gets a weird mix of quality.
- Models are really bad out of the box at humour.
At the end of the day it's just a hard problem that takes a lot of work and still isn't solved. GEPA prompts help, if you have good evals. Supervised fine-tuning works a little bit, but only if you training on a chain-of-thought thinking phase. We have a new evaluation builder that uses examples of edge cases for alignment, an
Read the rest
d jokes require the most iteration and feedback for refinement.
If you want to try it: https://github.com/kiln-ai/kiln
Kiln wraps up all the parts in on app. Just drag and drop in files. You can easily compare different configs on your dataset: extraction methods, embedding model, search method (BM25, hybrid, vector), etc.
It uses LanceDB and has dozens of different extraction/embedding models to choose from. It even has evals for checking retrieval accuracy, including automatically generating the eval dataset.
You can use its UI, or call the RAG via MCP.
https://github.com/kiln-ai/kiln
https://docs.kiln.tech/docs/documents-and-search-rag
In Ask HN: How are you doing RAG locally?Line 1 makes it pretty clear:
# The contents of this file are adapted from the promptwrite library (https://github.com/StacklokLabs/promptwright),
# which was adapted from the pluto library (https://github.com/redotvideo/pluto).
https://github.com/Kiln-AI/Kiln/blob/d38a64b598bf21939263bed...
Curious how the OP "just confirmed. No code copied."
In DeepFabric – Generate high-quality synthetic datasets at scaleShow HN: Kiln – AI Boilerplate with Evals, Fine-Tuning, Synthetic Data, and Git
I noticed there weren't boilerplates for AI projects like there were for web apps, so I built one. Same idea - everything you need to get a project up and running quickly. However, instead of web-framework/CSS/DB, it's tools for AI projects: evals, synthetic data gen, fine-tuning, and more.
Kiln is a free, open tool that gives you everything most AI projects need in one integrated package:
- Eval system: including LLM-as-judge evals, eval data generation, human baselines
- Fine-tuning: proxy to many fine-tuning providers like Fireworks/Together/OpenAI/Unsloth
- Synthetic data generation: deeply integrated into evals and fine-tuning
- Model routing: 12 providers including Ollama, OpenRouter, and more
- Git-based collaboration: projects are designed to be synced through your own git server
The key insight is that these tools work much bette
Read the rest
r when they're integrated. For example, the synthetic data generator knows whether you're creating data for evals vs. fine-tuning (which have very different data needs), and evals can automatically test different prompt/model/fine-tune combinations.
It runs entirely locally - your project data stays in local files, and you control your own git repos. No external services required (though it integrates with them if you want).
Main project GitHub: https://github.com/Kiln-AI/Kiln
Demo GitHub where I use it to build a 'natural language to ffmpeg command' demo with evals, fine-tunes, and synthetic data (including demo video): https://github.com/Kiln-AI/demos/blob/main/end_to_end_projec...
This is a bit confusing, I sort of expected it to be a bit like Kiln https://github.com/Kiln-AI/Kiln to generate datasets for AI, but it looks like the outputs are more just data / files than datasets?
In Show HN: I built an AI dataset generatorWe don't sell fine-tuning tools - we're an open tool for finding the best way of running your AI workload. We support evaluating/comparing a variety of methods: prompting, prompt generators (few shot, repairs), various models, and fine-tuning from 5 different providers.
The focus of the tool is that it lets you try them all, side by side, and easily evaluate the results. Fine-tuning is one tool in a tool chest, which often wins, but not always. You should use evals to pick the best option for you. This also sets you up to iterate (when you find bugs, want to change the product, or new models comes out).
Re:demo -- would you want a demo or detailed evals and open datasets (honest question)? Single-shot examples are hard to compare, but the benefits usually come out in evals at scale. I'm definitely open to making this. Open for suggestions
Read the rest
on what would be the most helpful (format and use case).
It's all on Github and free: https://github.com/kiln-ai/kiln
What’s the thinking of spreadsheet first? Just making it super accessible for people who already have data?
I’m building a UI for fine tuning (and evals, and synthetic data gen) - https://github.com/Kiln-AI/Kiln - and went the custom UI route. From chatting with folks - most people don’t have datasets, and need help building them.
In Show HN: Create your own finetuned AI model using Google SheetsShow HN: Kiln – GitHub+Jira+Figma for AI product development
I've been hard at work on Kiln (3k Github stars, ~10k users):
- Gihub: https://github.com/Kiln-AI/Kiln
- Demo Video: https://getkiln.ai#demo
It's goal is to makes AI model+product development processes better in a bunch of ways:
- Truly intuitive desktop apps, on top of our open-source python library
- Rapid prototyping: Quickly try a range of state of the art models+techniques (no code needed)
- Task specific evals: How does the system perform on your goals (not just global benchmarks)
- Iterate with confidence: update to better+faster models/techniques, knowing it works on all past evals/tests
- Team collaboration: get the entire team (PM/QA/Eng/Design/DS) engaged early and often. Shared datasets+evals you can sync with Git.
Features include:
- Fine tune dozens of models
- Interactive synthetic data generation to build train/eval datase
Read the rest
ts
- Powerful custom evals: LLM as Judge and G-Eval
- Collaboration: share with Git or shared drives
- Runs any LLM: locally or remote
- Private: it's local software
- Open Source Python library
- Completely free
- Much more: dataset management, ratings, prompt generators, JSON support, LLM repairs, reasoning models...
Happy to answer any questions!
Agreed if you are building a typical service. The abstractions will slow you down and don’t and anything.
The use case where they are helpful is “bring your own keys” apps. I maintain https://github.com/kiln-ai/kiln which allows you to bring keys for 13 different providers. The abstraction is very much worth it for me.
That said:
- I migrated from LangChain to LiteLLM and never looked back
- I have over 1000 automated integration tests that check the grid of LLM features (tools, json), model, and provider. Without them it would still be a mess.
In We chose LangGraph to build our coding agentIf you expand into tools for developing AI models, check out Kiln: https://github.com/Kiln-AI/Kiln
It includes synthetic data generation, fine-tuning and evals to help build your own models.
In Show HN: My Attempt to Organize the World of AI Dev ToolsShow HN: Distill your own model from DeepSeek R1 in 30 mins (open+GUI)
I've built a GUI tool that lets you distill R1 into a custom fine-tune. It includes synthetic dataset generation: creating a task specific dataset from R1.
The demo video shows building 4 R1 distilled models of different sizes: llama (3b, 8b, 70b) and GPT4o. The whole process took less than 30 minutes (not including training time).
- Guide with video: https://docs.getkiln.ai/docs/guide-train-a-reasoning-model
- Github: https://github.com/Kiln-AI/Kiln
- Download: https://github.com/Kiln-AI/Kiln/releases/tag/v0.10.0
4 points · Open discussionFine tuning to a specific codebase is a bit strange. It's going to learn some style/tool guidance which is good (but there are other ways of getting), at the risk of unlearning some generalization it learned from looking at 1,000,000x more code samples of varied styles.
In general I'd suggest trying this first:
- Large context: use large context models to load relevant files. It can pickup your style/tool choices fine this way without fine tuning. I'm usually manually inserting files into context, but a great RAG solution would be ideal.
- Project specific instructions (like .cursorrules): tell it specific things you want. I tell it preferred test tools/strategies/styles.
I am curious to see more detailed evals here, but the claims are too high level to really dive into.
In generally: I love fine tuning for more specific/repeatable tasks.
Read the rest
I even have my own fine-tuning platform (https://github.com/Kiln-AI/Kiln). However coding is very broad. Good use case for foundation large models with smart use of context.
Show HN: Kiln - Interactive LLM fine-tuning, dataset collab & synthetic data gen
Hi HN!
Excited to share a project I’ve been working on called Kiln AI! It solves what I’ve always found to be the hardest part of building AI products: products simply don’t have datasets, and datasets can’t keep up with product evolution. Product goals evolve, are hard to define, and new bugs emerge all the time. Kiln helps you build ML models for your product, from a collaborative dataset. For a bit of context: I’ve been building consumer AI products for a decade at Apple, my own startup, and MSFT.
At its core Kiln is 3 things (so far):
- Really great collaboration: Kiln datasets are designed to be shared/versioned in Git. We make it easy for the full team (PM/QA/subject-experts) to contribute directly, via super intuitive apps. With Kiln, when anyone on the team adds bugs/evals/goals, they go right into the dataset to be picked up in t
Read the rest
he next build.
- Rapid fine-tuning: dispatch fine-tuning jobs for a range of top models (Llama 3.2, Mixtral, GPT 4o/4o-mini). It’s fine tuning in just a few clicks.
- Synthetic data generation: our interactive data gen helps create large enough datasets for fine-tuning and evals. Build an initial training dataset, or use it to build out enough examples of a bug for the model fix it. It uses large models and heavy prompts (COT, multi-shot), which allows you to fine-tune smaller and faster models.
The link here is to the new fine-tuning feature which just launched today. The demo shows starting a project from scratch, defining a task, generating synthetic training data, fine-tuning 9 models, and deploying them. It’s all very easy: 18 mins of active work, all in an intuitive UI. You can download the apps and follow the same process for your own product goals.
I’d love to chat with folks building AI products, and offer any help I can (tools and/or guidance). Fire me a message at steve@getkiln.ai if interested.
You can download our apps for Mac, Windows or Linux, or `pip install kiln_ai` for the library.
Github with docs, downloads, and guides: https://github.com/Kiln-AI/Kiln
Show HN: Interactive synthetic data generation for LLM fine-tuning
2 points · Open discussion