MLJAR, Inc.

MLJAR Studio 1.2.1 available. What is new: - 1-click conversation sharing - 1-click sharing notebooks as web app with Mercury - better UX in get started view You can download at https://lnkd.in/dYEm9uvv

I added conversation sharing for AI data analyst chats inside MLJAR Studio. The release 1.2.1 is building right now, and will be available in few minutes ⏰ In the first conversation I asked it to load 7 day weather data for Łapy, Poland 🇵🇱 and compute forecast. And you know what AI did? It simply compute hourly average from previous days ... Anyway, the first conversation is public https://lnkd.in/ddJbUpzR You can check the details by opening the code blocks, I put a lot of ♥️ to make it look like this - feel free to expand code blocks to see Python code

🚀 MLJAR Studio has a free version ✅ We want to make Python data analysis easier to start. You don’t need to spend time on setup before you can explore your data. You can open a CSV file, write Python code, use no-code tools, and get help from an AI assistant in one place. The free version is a good way to try MLJAR Studio and see how it can help in daily data work. It is especially useful if you are: ✅ learning Python, ✅ starting with data analysis, ✅ working with CSV or Excel files, ✅ creating reports, ✅ exploring datasets, ✅ looking for a simple Python workspace. https://mljar.com/ Python is powerful, but the first steps should be simple. That’s why we made a free version of MLJAR Studio. 🙌 Try it and let us know what you think 🤔 .

Do you trust your AI in data analysis tasks? I asked AI to load CSV data. I used simple prompt `read a file`. The AI generated code, displayed header of loaded data, and run verification prompt. Thanks to this I can visually inspect result and there was additional LLM pass. Do you have human and AI in the loop for your data analysis tasks?

We realized that many AutoML tools are powerful… but still stressful for new users. So we expanded the MLJAR AutoML documentation and workflows inside MLJAR Studio. The goal is simple: make local machine learning easier to start, easier to understand, and easier to trust. Here’s how the workflow looks step by step ↓

🤖 Ready for the fresh update for AutoML opensource projects? If you haven’t looked at AutoML tools recently, the ecosystem may surprise you. A few years ago, most AutoML discussions focused on one thing: 𝘞𝘩𝘪𝘤𝘩 𝘭𝘪𝘣𝘳𝘢𝘳𝘺 𝘨𝘪𝘷𝘦𝘴 𝘵𝘩𝘦 𝘣𝘦𝘴𝘵 𝘣𝘦𝘯𝘤𝘩𝘮𝘢𝘳𝘬 𝘴𝘤𝘰𝘳𝘦? In 2026, the conversation is much broader. AutoML is no longer only about model selection. Read more on our blog: https://lnkd.in/dpuhSwHt article by Piotr Płoński

Do you still manually write code for statistical tests? Or just... ask AI to run them and explain the results? Here's an example of conversational AI doing t-test and ANOVA in a notebook: https://lnkd.in/dseKugru The gap between "writing stats code" and "understanding stats results" is finally closing.

We’ve just expanded the documentation for AutoML — and it’s now much more comprehensive 🚀 If you’re working with machine learning in Python, the updated docs walk you through the full workflow: from quick model training to more advanced topics like feature engineering, model tuning, ensembling, and interpretation. We’ve also added clearer explanations, practical examples, and guidance on how to get the most out of automated pipelines. 👉 Check it out here: https://lnkd.in/dSGwJDek As always, feedback is welcome — we’re continuously improving both the tool and the docs.