an investor platform that runs python in the browser

Built an interactive investor platform for a climate-adaptation venture fund. It replaces the static pitch deck with a model you can drive. Move the climate scenario and the market sizing, the demand multipliers, the fund returns, and the pipeline all recompute in front of you. The whole thing runs Python in the browser with no server behind it.

the idea it argues

The thesis is that climate adaptation is a budgeted, growing market rather than a hope. As warming pushes past 2 degrees, governments, insurers, and enterprises are forced to spend on resilience. The platform models that market expanding several-fold over the coming decades across a handful of sectors where demand rises directly with climate volatility.

the interactive core

The centerpiece is a climate scenario control. Set the warming level and the model applies a demand multiplier, because each additional half degree of warming raises adaptation spending. That single input cascades through the whole deck. Sector growth, total addressable market, and the fund math all respond to it.

The platform is organized as a set of investor views: the thesis, the pipeline, LP returns, fund modeling, impact, and the competitive position. Each one is a live panel rather than a screenshot, so a prospective investor can pressure-test the assumptions instead of trusting a fixed number.

how it is built

The part worth calling out: the model is written in Python and rendered with Panel, Bokeh, and Plotly, then compiled to WebAssembly and executed client-side through Pyodide. The financial logic, the charts, and the scenario engine all run on the page. There is no API call, no backend to maintain, and no spreadsheet to email around. You ship one static bundle and the recipient gets a working model.

why this approach

Investor materials usually force a choice between a polished deck that cannot compute and a spreadsheet that nobody else can open. Putting a real Python model behind a clean interface removes that tradeoff. The story stays interactive, the assumptions stay visible, and the analyst keeps the rigor of code while the audience gets something they can actually touch. That is a better way to make an argument with numbers.