The engine is now open source
Financial Model Engineering

Financial Models Should Be Engineered

Write models in Python, deliver as Excel with real formulas.

Versioned. Connected to data. AI‑readable. Human‑transparent.

The engine is now open source.

$ pip install modeleon

If Modeleon resonates — a GitHub ⭐ helps more people find it.

The problem

There is no proper tool for financial model automation yet.

>80%

of Excel models contain errors.

47%

of CFOs say correcting errors and chasing data impairs strategic decision-making.

0

version control. No deep dependency graph. No testing. Models live in one person's head and die when they leave.

89%

of finance teams still rely on Excel even when they have planning software.

The pattern

Every other data workflow is already engineered.Financial modeling isn’t.

Data ingestion, transformation, storage, visualization. All versioned, tested, automated. But the financial model, where the actual decisions happen, is still a spreadsheet someone emailed.

Data Transformation

Analytics Engineering

dbt

Versioned, tested, documented

Data teams version and test every SQL query. Financial models have no version history at all.

Infrastructure

Infrastructure as Code

Terraform

Automated, repeatable, auditable

DevOps teams describe servers in code and deploy with one command. Finance rebuilds the same model from scratch every quarter.

Financial Modeling

Financial Model Engineering

Modeleon

The most valuable layer in the stack

Where decisions actually happen. Still unengineered. It's time.

What we’re building

The first platform built for Financial Model Engineering.

✓ Engine open-sourced→ Platform coming soon

Python → Excel with real formulas

✓ Available now

Write revenue = price * volume in Python. Get =B4*B5 in the Excel cell. Not values. Live formulas.

AI that reads your logic

AI reasons about the model's dependency graph, not flat cells. Structured code is fundamentally easier for AI to understand than spreadsheets.

Version control, built in

Every model is code. Git-diffable. No more Q3_Budget_v7_FINAL_v2.xlsx.

Connected to any data

Models read from any source: data warehouses, ERPs, APIs, databases. No manual copy-paste.

Advanced analytics

Monte Carlo, optimization, time-series forecasting, machine learning. The full Python ecosystem inside your financial model.

Audit every number

Every output traces to a formula. Every formula traces to an assumption. Auditors, boards, and regulators can verify.

Financial models deserve engineering.

Read the full case for why financial models should be engineered. And why now.

Read the Manifesto

We’re building it.