AcademiaML

coming soon

I'm building AcademiaML as a notebook-first Electron desktop tool for researchers with tabular data who need help selecting and running ML workflows.

View source

Notebook context stays local before model guidance leaves the workspace.

Local Traceable Approved

Why I'm building it

I'm building this for researchers who want Jupyter-style notebooks, local Python environments, and advisor chat without giving up project ownership.

I'm shaping AcademiaML like a research workstation: real folders, real notebooks, local kernels, data summaries, and approval-based agent help.

Capabilities

Readable project folders with notebooks, raw data, derived data, artifacts, and logs.

Per-project Python runtime bootstrapping and local Jupyter kernel execution.

Advisor chat for OpenAI-compatible providers using only approved data context.

Agent queue and approvals for local Copilot-assisted workflow steps.

Workflow trace

1 Create or open a project folder.
2 Import tabular data and inspect inferred schema, column types, and summaries.
3 Work in notebooks while the advisor suggests modeling paths.
4 Approve broader agent actions only when the research context is ready.

Download

I'll publish signed downloads through GitHub Releases.

I've already configured this page for OS-aware release assets. Until signed binaries are published, the primary action stays disabled and the source remains available.

View source