Hi, I'm Lexi Ouellette

Property Manager Reporting & Insights

Overview

• Enhanced transparency and efficiency of rental deposit refund payments using data analytics.
• Analyzed anonymized datasets on disbursements, recipients, and properties to extract financial insights.
• Built Power BI reports for performance tracking, error detection, and identifying geo/demographic trends.
• Integrated Microsoft Fabric Lakehouse with SQL + Python to automate reporting and decision support.

Tech Stack

• Microsoft Fabric – Lakehouse storage and PySpark SQL transformation
• Power BI Desktop – Interactive dashboards, DAX, and visuals
• Python (matplotlib, seaborn) – Custom visualizations and correlation analysis
• DAX – KPI calculations, dynamic fields, and grouped comparisons

Data Sources

• Static Data: Disbursements, Recipient Payments, Feedback, Apartment Names
• Census Data: Income, Business activity, Unemployment, Education, Housing, Poverty, Family Demographics
• Static data was provided by the company to evaluate refund program effectiveness.
• Census data was collected to uncover external socioeconomic factors affecting completion rates.

Quantum Circuit Cutting

Overview

• We used wire cutting to split a 7-qubit circuit into two 4-qubit circuits, allowing broader circuit decomposition at the cost of added overhead.
• This method uses Move instructions—functionally like swap gates—to transfer quantum state information between the two circuit halves via an ancillary circuit.
• Each subcircuit is executed separately, once on a simulator and once on IBMQ’s quantum hardware.
• The resulting sub-experiment data is used to build Quasiprobability Distributions, which are then combined to reconstruct the expectation value of the original 7-qubit circuit.

Tech Stack

• Python – For writing quantum circuit logic, estimation functions, and handling data
• Qiskit – IBM's open-source quantum computing SDK used to – Build and cut quantum circuits, Execute on simulators (AerSimulator) and real quantum hardware (IBMQ backend), and Use Estimator for expectation value calculations
• IBM Quantum (IBMQ) – Cloud-accessible quantum hardware for running real quantum circuits
• Qiskit Runtime / Estimator V2 – Used as the interactive environment to write, run, and visualize the experiments
• Google Colab + Jupyter Notebook – Used as the interactive environment to write, run, and visualize the experiments
• NumPy - For numerical operations such as rounding and calculating relative errors
• Matplotlib - For visualizing outputs

Simple Protocol

Overview

• This project implements a basic TCP-based client-server protocol for controlling a virtual light.
• The protocol uses structured headers and logs all communication between a client and server to log files.
• Communication is based on a simple protocol:
• All messages start with a 12-byte header: version (4 bytes, integer), msg_type (4 bytes, integer), msg_length (4 bytes, integer)
• This is followed by a UTF-8 encoded message body<
• Message Types:
• 0: HELLO message
• 1: LIGHTON command
• 2: LIGHTOFF command

Tech Stack

• Python - Main implementation of client-server communication
• socket module - TCP/IP-based socket programming
• struct module - Packs/unpacks custom binary headers
• logging module - Logs events, messages, and errors to console and file
• argparse - command-line arguments for client/server setup
• pcap - Analyze packets exchanged over the network

Wedding Planner Algorithms

Overview

• This project includes two tools for planning a wedding or large event:
• 1. Seating Chart GeneratorAutomatically assigns guests to tables based on their tag (bride_family, groom_family, couple, friend), keeping couples together and grouping families and friends logically.
• 2. Table Calculator
Calculates all possible table configurations (round, rectangle, or mixed) to accommodate a given number of guests. Includes min/max seating ranges and extra values.

Tech Stack

• Python

Wedding Planner Web Application

Overview

• An interactive wedding seating chart planner built with React, React-Konva, and Bootstrap on the frontend, and FastAPI and JSON on the backend.
• This tool allows users to manage guests, generate seating charts, create floor plans, and calculate tables/seats based on guest count.
• Ingrated the wedding planner algorithms into a user friendly web application.

Tech Stack

• React - Frontend framework
• React-Konva - Canvas rendering
• Bootstap - UI components
• Axios - Frontend API requests
• FastAPI (Python) - Backend logic and building APIs
• JSON - Storage
• CORS - Security

Features

• Upload or manually enter guest lists (with tags and groups)
• Automatically calculate table arrangements based on guest count
• Create round or rectangle tables with configurable seat counts
• Visual floor plan canvas powered by React-Konva
• Modal editor for table properties and guest assignments
• Persist and load saved layouts using FastAPI