I am a CPA who spent years auditing public real estate investment trusts (REITs), developers, and investment funds at KPMG.
If you have ever tried to model a multi-tier commercial real estate capital stack or S-curve construction draw in Excel, you know the pain: time-consuming, fragile formula chains, hidden circular references, slow spreadsheet bloat when shifting exit cap rates, and constant fear that a single broken cell will corrupt the entire waterfall.
I wanted something deterministic, fast, and robust that runs locally without browser lag or spreadsheet risk. So, I built CREDevSim, a native desktop app for commercial real estate underwriting and waterfall simulations.
Why Godot?
An unusual choice for this: I built it using Godot Engine and GDScript. While Godot is known for game development, its node-based UI system (Control nodes, custom themes, instant layout scaling) made it remarkably fast and clean for building a high-density, responsive desktop data interface without dealing with Electron's heavy memory footprint or web browser limitations.
What it handles locally:
- Open it, enter a deal, done — zero setup, zero learning curve, zero configurations
- Local execution with zero browser lag or hidden formula errors
- Purpose-built for LP/GP waterfall mechanics, capital calls, and promote structures
It is currently in public beta as a desktop app, and I am looking for technical feedback, architectural critiques, or thoughts on the underwriting logic from anyone who regularly wrangles financial models or builds desktop utilities.
Would love to hear your thoughts or answer any questions!
I am a CPA who spent years auditing public real estate investment trusts (REITs), developers, and investment funds at KPMG.
If you have ever tried to model a multi-tier commercial real estate capital stack or S-curve construction draw in Excel, you know the pain: time-consuming, fragile formula chains, hidden circular references, slow spreadsheet bloat when shifting exit cap rates, and constant fear that a single broken cell will corrupt the entire waterfall.
I wanted something deterministic, fast, and robust that runs locally without browser lag or spreadsheet risk. So, I built CREDevSim, a native desktop app for commercial real estate underwriting and waterfall simulations.
Why Godot? An unusual choice for this: I built it using Godot Engine and GDScript. While Godot is known for game development, its node-based UI system (Control nodes, custom themes, instant layout scaling) made it remarkably fast and clean for building a high-density, responsive desktop data interface without dealing with Electron's heavy memory footprint or web browser limitations.
What it handles locally: - Open it, enter a deal, done — zero setup, zero learning curve, zero configurations - Local execution with zero browser lag or hidden formula errors - Purpose-built for LP/GP waterfall mechanics, capital calls, and promote structures
It is currently in public beta as a desktop app, and I am looking for technical feedback, architectural critiques, or thoughts on the underwriting logic from anyone who regularly wrangles financial models or builds desktop utilities.
Would love to hear your thoughts or answer any questions!