MLJAR Studio: Local AI Data Analyst That Generates Reproducible Notebooks

MLJAR Studio brings a local AI data analyst to your desktop. You ask questions in plain English, the AI generates Python code, runs it locally on your machine, and saves the entire conversation as a reproducible *.ipynb file. No cloud, no data egress – everything runs on your own computer.
Key Features
- Local environment: Sets up a Python environment automatically on Mac, Windows, and Linux, including installing missing packages on the fly.
- AutoML for tabular data: Built-in support for classification, regression, and multiclass tasks using the
mljar-supervisedlibrary. - Standard Python stack: Works with pandas, matplotlib, and other common libraries.
- File formats: CSV, Excel, Stata, Parquet, and more.
- Database connections: PostgreSQL, MySQL, SQL Server, Snowflake, Databricks, and Supabase.
- AI backends: Use Ollama locally (zero data egress), bring your own OpenAI key, or use the MLJAR AI add-on.
- Conversational notebooks: Every interaction is recorded as code in a notebook you can inspect, modify, and rerun.
- One-click web apps: Convert notebooks into interactive web applications using Mercury (open-source), self-hosted.
Pricing and Trial
$199 one-time payment with a 7-day free trial. No cloud services required – it's a desktop app that runs entirely locally.
Who It's For
Data analysts, data scientists, and researchers who need privacy, reproducibility, and control when using AI for data analysis – especially those working with sensitive data or wanting to avoid cloud dependencies.
📖 Read the full source: HN AI Agents
👀 See Also

sourcecode: Open-Source CLI to Compress Large Java/Spring Monorepos for Claude
sourcecode CLI reduces a ~4k-file Java/Spring monorepo from ~3M tokens to 1.7k tokens (compact mode). Currently focuses on context compression, git hotspot detection, and symbol lookup.

Local MCP Memory System with Consolidation for AI Conversations
A developer built an MCP server that provides persistent local memory for AI clients, using Qwen 2.5-7B to consolidate conversations into structured knowledge documents every 6 hours. The system runs entirely on your hardware with semantic dedup, adaptive scoring, and FAISS vector search.

Phantom: A Persistent AI Agent Built with Claude's Agent SDK
Phantom is an open-source Bun/TypeScript process that wraps Claude's Agent SDK (Opus 4.6) with persistent vector memory, a self-evolution engine, and an MCP server interface. It runs continuously on its own VM or Docker Compose and communicates via Slack.

AutoBe: How Weak Local LLMs Fixed an AI Backend Generator's Architecture
AutoBe is an open-source AI agent that generates complete backend apps using TypeScript, NestJS, and Prisma. The team discovered their initial 100% compilation success produced unmaintainable code, then rebuilt with modular generation—crashing success to 40%—and used weak local LLMs like qwen3-30b-a3b-thinking to debug schema ambiguities.