
I'm a data scientist who spent years building forecasting models. The biggest lesson: single-point forecasts are almost useless for long-horizon structural questions. What matters is the distribution.
So over the past year I built WorldSim. It runs Monte Carlo simulations for 195 countries across multiple socio-economic indicators, producing full P10/P50/P90 probability distributions instead of single estimates. 100+ coupling rules connect variables causally, so shocks cascade realistically.
Stack: Django, PostgreSQL, Celery, Stripe. Deployed on Hetzner.
Free tier available (3 runs).
Would love feedback from anyone in the macro/economics space. Try the country you live in: worldsimlab.com
Really interesting Monte Carlo simulations at a macro level can get complex quickly. I’m an app tester and I focus on testing usability, workflow clarity, and edge cases in data-driven platforms. I’d be happy to review it and provide structured feedback if you’re open to early testers