Lefts – a domain specific language for building creative ML models
Lefts is a small domain specific language for applied machine learning modelling. It is aimed at anyone that builds predictive models for a living and wants to focus on reasoning about model behaviour and building creative architectures, and not on building train/test pipelines or worrying about data leakage. It is simple but quite powerful - I have been using it in my own work to explore new ways of modelling (check out the tutorial on geometric models!), and to breeze past the least interesting parts of being a machine learning engineer. It also has some cool functional programming…
What it does
Composable transformations for ML model experimentation
Lefts is a very simple domain specific language for building complex machine learning workflows from simple ones. Starting with your favourite machine learning models, you can use Lefts operations to: Without making subsequent model fitting, evaluation or experimentation any more complex than it was with the original model. This implementation is built on top of the excellent Polars DataFrame library. A Lefts command creates a new model by transforming these functions into a new .fit and .predict. Because this new model also has a .fit and .predict, it can be transformed with further Lefts commands. The following code creates a rolling monthly retrain workflow: twelve copies of a Ridge…from nsmat.github.io
In the maker’s words, at launch
Lefts is a small domain specific language for applied machine learning modelling. It is aimed at anyone that builds predictive models for a living and wants to focus on reasoning about model behaviour and building creative architectures, and not on building train/test pipelines or worrying about data leakage. It is simple but quite powerful - I have been using it in my own work to explore new ways of modelling (check out the tutorial on geometric models!), and to breeze past the least interesting parts of being a machine learning engineer. It also has some cool functional programming going on under the hood (check out the design philosophy!).
Does the same job
all alternatives →- CTCedille, the largest French language model, released in open source2021 · ▲240
We are excited to announce Cedille, the largest language model for French (6b parameters). Demo: https://cedille.ai Language models are general purpose AI systems that are able to solve a range of tasks by simply being prompted for it. It can be used for example to summarize text, do translations, or for idea generation & overcoming writer's block. You may know GPT-3, the humongous model from OpenAI. Cedille is a similar model targeting the French demographic - but smaller, as we don’t yet have $1b in the bank like they do. Although GPT-3 supports multiple languages including…
- LILanguageModels.io2018 · languagemodels.io · ▲9
- LALLM Alignment Template – Aligning Language Models with Human Feedback2024 · github.com · ▲9
Hey Hacker News! I've been working on an open-source project called LLM Alignment Template, a comprehensive toolkit designed to help researchers, developers, and data scientists align large language models (LLMs) with human values using Reinforcement Learning from Human Feedback (RLHF). What the project does: Interactive Web Interface: Easily train models, visualize alignment metrics, and manage alignment with an accessible UI. Training with RLHF: Align models effectively to human preferences using feedback loops. Explainability: Built-in dashboards to help understand model behavior using…
- DADeepClause – A Neurosymbolic AI System Built on WASM and PrologNov 2025 · github.com · ▲7
Hi HN, Today I'd like to present the results of my weekend project of the last year or so. Given there are many posts on HN about LLMs and Prolog, I thought that this would be of interest. DeepClause is my own (possibly misguided :-) attempt at combining LLMs with Logic Programming, ultimately hoping to establish a foundation for building more reliable agents, that produce reproducible and fully traceable result. At the heart of DeepClause is a DSL called "DeepClause Meta Language" (DML) which can be used to encode agent behaviors as executable logic programs. DML is executed by a…
- WBWe built the Keras of LLMs2025 · github.com · ▲5
Here is a production-first Keras-inspired LM framework, built with the advice of François Chollet (ex-Google, creator of Keras and ARC-AGI), our technical advisor. This system have already been deployed in production with our clients (which is why we have already every LLMOps practice implemented). It is also compatible with Jupyter and Marimo to integrate seamlessly in you Data Scientists workflows. You can try the code examples online on HF space and you can find more information in the documentation and FAQ. If you have any feedback for us don't hesitate to join our discord! More releases…
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, August 2026
the whole month →- TL
Life & fun · 10d ago · louisabraham.github.io
- SA
Hello HN! I found that picking out plausible but diverse skin tones for my digital art and game development projects was kind of difficult, and I got curious about if there was a way to define a color space that made it easy. I've built a color picker and procedural generation algorithm based on the space as well as a bunch of other fun js features and demos throughout the page that use the equations. If you find it interesting, I have lots of explanations of how I built it and what properties the space has. The methodology might be a bit shaky, but hopefully the result is as helpful for…
Life & fun · Aug 2026 · toneyalexander.github.io
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com