Visual data science,
block by block.
Build, iterate, and share reproducible pipelines without the boilerplate.
Why Forge
See your pipeline, not just your code
Visual blocks make data science approachable. Drag, connect, run.
Edit a parameter. Run only what changed.
Targeted re-execution — not full notebook reruns. Iterate in seconds, not minutes.
Let AI agents do data science through Forge's MCP
Claude Code, Codex, and other AI agents can build, run, and interpret pipelines programmatically.
Three steps to your first pipeline.
Drag blocks onto the canvas
Pick from 70+ built-in blocks or write your own Python. Each block is a self-contained step.
Connect them into a pipeline
Draw edges between blocks to define data flow. Forge handles dependencies automatically.
Run & iterate
Change a parameter, hit "Run Stale" — only downstream blocks re-execute. No more full reruns.
From quick experiments to production pipelines.
Data cleaning & transformation
Wrangle messy CSVs into structured datasets with reusable blocks.
ML model training & evaluation
Train, tune, and compare models with visible data flow at every step.
Exploratory data analysis
Prototype fast. See intermediate results without re-running everything.
Reproducible research workflows
Save and share pipelines as JSON. Same inputs, same outputs, every time.
Custom Python blocks
Extend Forge with your own code. Full Python - no sandbox, no limits.
Visualization & plotting
Scatter plots, heatmaps, bar charts, histograms - 11 built-in plot types, plus 3D.
Free & open source.
Forge is free and open source under the Apache 2.0 license. Built by a data scientist, for data scientists.
Premium features (team collaboration, cloud execution) coming soon.