Skip to content

Instantly share code, notes, and snippets.

View Kvit's full-sized avatar

Vitali Khvatkov Kvit

View GitHub Profile
@Kvit
Kvit / ClaimRecoveryAgent.md
Created October 31, 2025 17:04
Capability Overview of Claim Recovery Agent

Recovr - Claim Recovery Agent

The Claim Recovery Agent is a premium service of the Recovr product by Pinnacle Services Corporation. It is an AI-powered system designed to automate the recovery of medical claim denials. It can identify the root cause of a denial, determine the recovery steps, and complete actions to resolve denials for a given claim.

Capabilities Overview

Information Sources and Data Tools

The agent has real-time access to the following information sources and data tools for analysis:

  1. Claim and Remittance Advice
  2. Patient Billing History
@Kvit
Kvit / table_demo.py
Last active July 6, 2025 12:49
Example of Interactive table with FastHTML Datastar and GreatTables
############################################################################################
# REACTIVE TABLE DEMO WITH FASTHTML AND DATASTAR
#
# This example demonstrates:
# 1. Building a reactive data table with real-time filtering and pagination
# 2. Using Datastar for reactive UI updates without page refreshes
# 3. Implementing DaisyUI and TailwindCSS for modern styling
# 4. Handling user interactions with side effects (selection, filtering, pagination)
############################################################################################
@Kvit
Kvit / main.py
Last active March 20, 2025 15:45 — forked from axelknock/main.py
Example of Datastar with FastHTML, ported from a FastAPI example
import asyncio
import json
from datetime import datetime
from fasthtml.common import *
from fasthtml.starlette import StreamingResponse
from fasthtml.core import to_xml
from datastar_py.sse import SSE_HEADERS, ServerSentEventGenerator
@Kvit
Kvit / socket_test.py
Last active January 29, 2025 19:50
Trying to change session value in socket
# Testing managing session values with websockets in FastHTML
from fasthtml.common import *
import uuid
app = FastHTML(exts='ws')
rt = app.route
@rt("/")
@Kvit
Kvit / recursive-search-json.js
Last active April 18, 2021 04:45
Recursive search of complex JSON structures for a given key
const jsonSearch = (obj, searchKey, results = []) => {
const r = results;
// if this is array, iterate elements
if (Array.isArray(obj)) {
//console.log("- ARRAY LENGTH ", obj.length);
for (const element of obj) {
// console.log("-- searching array element: ");
jsonSearch(element, searchKey, r);
}
@Kvit
Kvit / conda setup digital ocean shiny.txt
Last active April 8, 2021 18:10
Setting up python conda environment for R on Digital Ocean for Shiny Apps
## In shell (under root user): Run R as shyny user
runuser -l shiny -c 'R'
## in R session:
#istall reticulate
if (! require(reticulate)) install.packages('reticulate', dependencies = TRUE)
# init Python
reticulate::py_available(initialize = T)
@Kvit
Kvit / pseudo_rules.txt
Last active May 26, 2020 16:15
Pseudo Billing Rules
global.ruleArea = ORDER
// basic rules
if (orderPayer == INSURANCE_PAYER ) { billTo = PRIMARY_INSURANCE; priceList = PRICE_INSURANCE }
if (orderPayer == CLIENT) { billTo = CLIENT , priceList = LookupPriceList (CLIENT) }
if (orderPayer == PATIENT) { billTo = PATIENT , priceList = PRICE_SELF_PAY }
if (orderPayer == null) stop (404, "Payer Not Found", @biller)
// client rules
if (orderClient $in [...] ) billTo = CLIENT
@Kvit
Kvit / nuxt-install.txt
Last active September 9, 2019 17:20
Installation Nuxt Vuitify FireStore
npm install -g @vue/cli
npm init nuxt-app <project-name>
yarn lint --fix
// firebase
https://medium.com/@pascalluther/nuxt-js-v2-firestore-ssr-938d8fb7d2b0
// linting
'npm install eslint babel-eslint eslint-config-prettier eslint-plugin-prettier eslint-plugin-vue eslint-loader prettier -D'
@Kvit
Kvit / tips.R
Last active December 9, 2020 07:07
R tricks and tips
# convert NA to Zero
log_pay_dt[,lapply(.SD,function(x){ifelse(is.na(x),0,x)}) ]
# convert NA to Zero for numberic cols only
cols = which( sapply(log_pay_dt, class) == "numeric" ) %>% names()
log_pay_dt[, (cols) := lapply( .SD, function(x){ifelse(is.na(x),0,x)} ), .SDcols = cols ]
# factor to character
col_class<-sapply(cpt_blend_summary, class)
@Kvit
Kvit / test.R
Last active March 10, 2017 07:48
R Test
x<-1:10