Carson Sievert cpsievert

Shiny for Python Loan Analysis Dashboard - Snowflake

A Python Shiny web application for analyzing loan data with interactive visualizations and state-level statistics.

This dashboard features an LLM chat sidebar powered by querychat that allows the viewer to ask questions to dynamically generate SQL sent to Snowflake using ibis and SQLAlchemy to find results.

Overview

This dashboard provides an interactive interface to explore and analyze Lending Club loan data. It features:

Session Persistence: Eliminating the Gray Screen of Death

Problem

When a Shiny app's WebSocket connection drops — whether from a network blip, laptop sleep, server error, or any other cause — the user sees a semi-transparent gray overlay with no recovery path (the "gray screen of death"). The session is permanently destroyed server-side, and all state is lost.

An existing session$allowReconnect() mechanism exists but is off by default, only works on certain hosting platforms, and performs a cold restart (new session, server function re-runs, all server-side state lost).

Goal

This gist contains a minimal reprex for this CS ticket

The app.py was deployed here. If you visit there, and repeatedly refresh, you should (eventually) see this error:

Note that the reprex needs this branch of shinywidgets, which switches the order of the libembed-amd.js and output.js scripts in the head of the document.

For some reason, quarto serve seems to be ordering these `

	from chatlas import ChatAnthropic
	from faicons import icon_svg
	from shiny import reactive
	from shiny.express import input, render, ui

	chat_model = ChatAnthropic(
	system_prompt="""
	You are a helpful AI fitness coach.
	Give detailed workout plans to users based on their fitness goals and experience level.
	Before getting into details, give a brief introduction to the workout plan.

	import numpy as np
	from chatlas import ChatAnthropic
	from sentence_transformers import SentenceTransformer

	# Initialize the SentenceTransformer model and chat client
	embed_model = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")
	chat = ChatAnthropic(
	system_prompt="""
	You are a helpful AI assistant. Using the provided context,
	answer the user's question. If you cannot answer the question based on the

	import pandas as pd
	import plotly.express as px
	import plotly.graph_objects as go
	from shinywidgets import render_plotly

	from shiny import reactive
	from shiny.express import render

	num_rows = 10
	x_points = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]

	# Example adapted from https://plotly.com/python/mapbox-county-choropleth/

	import json
	from urllib.request import urlopen

	import pandas as pd
	import plotly.express as px
	import plotly.graph_objects as go
	from shinywidgets import render_plotly

	# ------------------------------------------------------------------------
	# Learn more about bslib::tooltip() and bslib::popover() at
	# https://rstudio.github.io/bslib/articles/tooltips-popovers/index.html
	# ------------------------------------------------------------------------

	library(shiny)
	library(bslib)
	library(palmerpenguins)
	library(ggplot2)

	library(bslib)
	library(htmltools)
	library(crosstalk)
	library(plotly)
	library(leaflet)

	# Create Diamonds view
	dat <- SharedData$new(dplyr::sample_n(diamonds, 1000))

	sidebar_diamonds <- layout_sidebar(

	library(bslib)
	library(bsicons)
	library(crosstalk)
	library(plotly)

	dat <- SharedData$new(dplyr::sample_n(diamonds, 1000))

	accordion_filters <- accordion(
	accordion_panel(
	"Dropdowns", icon = bs_icon("menu-app"),