Skip to content

Finance · Analytics · Automation

Data Powered
Solutions

Empowering businesses with automated financial insight; faster, clearer, smarter.

Contact MeSee the work
Deferred revenue roll-forwardSample model · USD 000s
01.1K2.2KOPEN1,240+ NEW860− REC(980)± ADJ(140)CLOSE980
Toolkit
  • Python
  • SQL
  • Alteryx
  • Power BI
  • Tableau
  • Agentic AI

Selected work

Portfolio

Nine end-to-end builds, from the SQL, Python, and Alteryx that shape the data to the dashboards people actually open. Most are explorable here in the browser.

Live model

React · Financial modeling

Backlog and Percentage-of-Completion Forecast

A driver-based forecast for a project business recognizing revenue over time: bookings, completion rate, cost-to-complete drift, and margin feed a quarterly backlog roll-forward, with scenarios and a two-way sensitivity grid. The arithmetic runs in the browser, so the model can be interrogated rather than read.

  • Scenario switching
  • Two-way sensitivity
  • Backlog roll-forward
Case study →
Real-world model

Power BI · MySQL

Finance Intelligence Suite

ASC 606-compliant revenue tracking, customer segmentation, and payment insights, built on an ERP-shaped schema staged in MySQL and transformed via SQL views. Users can explore contract-level trends and milestone obligations for revenue recognition analysis.

  • ASC 606
  • Subscription vs product revenue
  • Contract drilldown
  • Billing analysis
Case study →
Real-world model

AWS Bedrock · Python · SQL · React

AI Finance Data Agent

A conversational agent for financial analysis: a question asked in plain English returns a written answer together with the chart and the underlying numbers it was drawn from. Built on AWS Bedrock over a SQL data layer, with suggested follow-ups that carry an analysis past its first answer.

  • Natural language queries
  • Answers with evidence
  • Guided follow-ups
Case study →
Real-world model

Alteryx · Python

Alteryx Workflow Automation

Two Alteryx workflows behind recurring forecast and backlog reporting. Data is pulled from database and spreadsheet sources, cleansed and matched, reconciled against actuals, and written back out, replacing a manual monthly assembly process.

  • Multi-source inputs
  • Data cleansing
  • Process automation
Case study →
Real-world model

SQL · Power BI

Data Preparation Using SQL

This project models data staged in a relational database into conformed dimensions and two fact streams (Milestones, Subscriptions), exposed as SQL views for Power BI. It supports ASC 606-compliant revenue recognition, contract-level drilldowns, segment slicing, cost attribution, and payment timing analysis.

  • 5 SQL views
  • 4 conformed dims
  • 2 fact streams
Case study →
Real-world model

Power BI · SQL

Executive Summary

A high-level KPI overview tailored for upper management, summarizing business health using revenue, cost, conversion rate, and active customers. Users may explore performance obligations as needed. Prepared and cleaned in SQL, then presented in Power BI.

  • Customer profitability
  • Obligation drilldown
  • Management-ready view
Case study →
Real-world model

Power BI · SQL

Customer Churn & Retention Analysis

Active customers, CLTV, churn rate, and average satisfaction rating, with churn broken down by service type and customer reason. The dataset is prepared in SQL and visualized in Power BI.

  • CLTV
  • Churn drivers
  • Segment slicing
Case study →
Public dataset

Power BI · Python · Natural Language Processing

Customer Review and Sentiment Analysis

Using the Yelp Open Dataset, this project applies a fine-tuned roBERTa sentiment model to help with customer review analysis. The dataset was prepared and explored using Python and visualized in Power BI.

  • AI sentiment analysis
  • Python data prep
  • Geospatial analysis
Case study →
Public dataset

Python · Parquet

Data Preparation Using Python

This project integrates Yelp review, business, and user datasets for the Greater Philadelphia area, applies category standardization, and filters to relevant business types. Data cleaning, category normalization, and spatial filtering produce a refined Parquet dataset for focused regional analysis.

  • 3 datasets joined
  • Spatial filtering
  • Parquet output
Case study →