HN Showcase
Back to browse
GitHub Repository

Probabilistic Spreadsheets

1 starsTypeScript

Spreadsheets Cells with Uncertainty Distributions

by tobiadefami·May 13, 2026·4 points·0 comments
Visit ProjectView on HN

AI Analysis

●●●BangerBig BrainNiche Gem

Monte Carlo sampling inside a spreadsheet beats collapsing uncertainty into single scalars.

Strengths
  • •GaussFormula engine modifies HyperFormula to support distribution types natively.
  • •Confidence interval notation like N.CI(10, 20) makes uncertainty explicit in cells.
  • •Dependency graph view traces how uncertainty propagates through formula chains.
Weaknesses
  • •Early stage with only 182 commits; lacks advanced charting or export options.
  • •Requires Bun runtime, limiting adoption for teams standardized on Node.
Category
Data
Target Audience

Data scientists, risk analysts, quantitative modelers

Similar To

Guesstimate · Causal · Crystal Ball

Post Description

Hi HN,

I built a spreadsheet for modeling uncertainty directly in cells.

Cells can contain probability distributions, ranges, or distributions inferred from confidence intervals. Formulas then propagate that uncertainty through the sheet.

For example:

A1 = N.CI(10, 20)

B1 = U(2, 5)

C1 = A1 * B1

Instead of a single value, C1 becomes a sampled result.

The formula engine is powered by GaussFormula, a modification of HyperFormula that supports uncertain cells and sampled calculations.

The app is early and under active development and I would appreciate some feedback.

Similar Projects