Allie Bergmann

St.Louis, MO 63119 · (314) 599-7060 · 11arbergmann@gmail.com

A data scientist with 7 years of quantitative research, AI, and analytics experience in both the trading and business intelligence space. Experience with large data sets and developing data-driven insights for early-stage fintech and Fortune 500 companies. Intending to continue on my career path to leverage data to drive business decisions and innovative solutions, preferably in the BI, Virtual Reality, or trading domain.


Professional Experience

Sisu Ventures, LLC

Owner/Founder
  • A passion project dedicated to leveraging AI in proprietary innovation initiatives, with a primary focus on the fintech space.
  • Build and maintain algorithmic trading pipelines and dashboard performance monitoring, with fully autonomous rebalancing and risk management controls. This also includes using standard Python ML libraries, administering PostgreSQL databases, and maintaining GCP virtual machines in a customized ecosystem.
  • Offer consulting services in the Data Analytics/Data Science domains, with a focus on business intelligence applications.
Jan 2024 - Present

Ulta Beauty

Lead Data Scientist - Digital Innovation
  • Manage a team of data scientists with a proven track record of developing talent, building high-performing data-focused teams, and facilitating cross-disciplinary projects while fostering organizational collaboration.
  • Design, research, and implement data science projects that leverage computer vision, gamification, AI, and GenAI in the retail digital innovation space.
  • Rebuild and improve legacy recommender systems and customer-facing personalization integrations in digital channels.
  • Lead an end-to-end wayfinding indoor positioning initiative pilot and facilitated third party vendor relationships.
  • Lead architectural restructuring initiatives, such as VM migration for the team to updated GCP ecosystem.
  • Apply a spectrum of data science methodologies to address real-world business problems and requests to enhance product value and customer experience.
Development / Training
  • New Leader Development Program: A structured leadership development program which includes networking, assessments, and formal and informal leadership learning experiences. A three-day experience with a focus on "Building Your Talent Bench," "Developing Your Associates," and "Strengthening Your Leadership Capabilities."
June 2023 - Present

EXEGY

Data Scientist - Quantitative Research
  • Design and research statistical arbitrage strategies and end-to-end pipelines for intraday and daily trading, specializing in the momentum trading space.
  • Wrangled intraday trading datasets of multiple petabytes in size for use in statistical models.
  • Refactor and redesign research tools and pipelines, such as dynamic strategy back testers.
  • Refresh and retrain existing machine learning models for client-facing products.
December 2022 - April 2023

EDWARD JONES

Data Scientist
  • Lead efforts on financial advisor clustering analysis to evaluate FA P&L efficiency and determine FA professional objectives.
  • Partnered with dynamic portfolio optimization team to lend strategic and technical support on a tool that leverages trade scheduling and portfolio rebalancing.
  • Led efforts on privileged projects in the AML space.
  • Developed sensitivity analysis, model performance analysis, and performed improvements for a suite of AML models.
  • Performed unit testing on internally created Python libraries.
Data Analyst
  • Led annual pricing review initiatives to benchmark company pricing policies and generated scenario analyses for future pricing changes. This effort led to a nearly year-long evaluation of present and future/long-term pricing strategies.
  • Created a dynamic practice impact calculator to determine changing industry effects on branches for varying business strategies. The calculator was leveraged for the production of a branch-facing tool to provide significant value to the field. This project utilized analytical acumen as it pertains to database capacity, best practices for calculation logics, and branch practice business acumen.
  • Worked with business owners and project leaders to develop KPIs, and leverage economic models in efforts to research new business strategy proposals.
  • Consulted on a DEI initiative design and analysis for financial advisor groups. This included conducting propensity scoring, control group design, A/B testing, and subsequent initiative efficacy analysis in Python.
  • Assisted in the review of a previous focus client measure and its associated supervised learning model to ensure continued feature relevance and potential use cases in other future business developments. This review also included the exploration of gaps and opportunities for the model going forward, and any potential road blocks in potential analyses. This project required significant ETL capabilities and data source acumen.
  • Peer reviewed a Letter of Authorization / fraud detection initiative that generated a significant monetary influence.
  • Assisted with development, scenario modeling, and peer review of a program migration model for potential future business segments, where over $1T of client funds and approximately 19,000 advisors would be affected by program parameters.
  • Facilitated supporting redundancies in Python for a supervised and unsupervised learning client lifetime value project generated in R, for future business purposes.
  • Developed a workbook library for basic Python training, as well as collaborated with the Data Science team on repositories containing code snippets and parameters for best practices.
February 2020 - December 2022

FEFA, LLC.

Director of Asset Management
  • Oversaw all operations, portfolio management, compliance procedures, and quantitative analytics for FEFA Asset Management.
  • Collaborated with senior management team to develop and adapt sales processes, company policies, and address any personnel concerns.
Assistant Director of Asset Management, Senior Investment Analyst
  • Created process for researching, building, and managing all 30+ company portfolios.
  • Automated asset management processes, including account information audits, form submissions, portfolio tracking and management.
Retirement Strategist
  • Led workshops and one-on-one retirement prep counseling sessions for federal employees.
March 2018 - January 2020

Salarium Trading, LLC.

Co-Founder, Head Trader
  • Co-founded a proprietary, algorithmically-based trading LLC specializing in swing trading strategies for U.S. equities and cryptocurrencies.
  • Built strategies, databases, risk management systems, self-balancing portfolio algorithms, and trading architectures from scratch using Python.
October 2017 - November 2018

Tradebot Systems, Inc.

Quantitative Researcher
  • Devised an indicator for use in a machine learning trading model that tracked spread deltas on a quote-by-quote basis to determine microstructure momentum.
  • Designed an exit strategy based on an alpha decay model to improve capture rate.
  • Developed a Jupyter library of technical analysis indicators and corresponding optimal parameters strategies.
July 2017 - September 2017

Kennedy Capital Management, Inc.

Equity Research Intern
  • Conducted fundamental research analysis and constructed financial earnings models to present buy/sell/hold recommendations with regard to stock purchase decisions on current and proposed positions.
  • Contacted public corporations and participated in meetings with senior management including CEO’s and CFO’s to obtain/update company database information.
  • Prepared evening reports which included positions held, cash balances, and daily customer account and employee performance.
Summer 2016

Education

Master of Applied Data Science

University of Michigan, School of Information
  • Inaugural Cohort
  • Pioneer Scholarship
  • Finance Data Science Club
September 2019 - December 2021


Master of Business Administration

University of Missouri
  • Graduate Teaching Assistant, Department and Sales Program
  • John Sublett Logan MBA Scholarship
  • Trulaske College of Business Crosby MBA Toyota Case Competition Winner
September 2015 - May 2017


Bachelor of Arts in Architecture

University of Kansas, School of Architecture, Design, and Planning
  • Monsters of Design Award, Contract Inspirations Award for Practice, AIA Kansas Excellence in Architecture Design Award, AIA Kansas Concept Honor Award
  • Study Abroad in Florence, Italy (Class focus in Language and Arts)
September 2011 - August 2014

Skills & Certifications

Certifications

Languages / Programs
  • Python: pandas | numpy | scikit-learn | statsmodels | matplotlib | pySpark | networkx | tensorflow | opencv
  • Databases: SQL: PostgreSQL, MySQL, Oracle, BigQuery | NoSQL: MongoDB, BigTable
  • Deployment: Docker | Kubernetes | Jenkins
  • Version Control: Git | Github | BitBucket
  • APIs: REST | gRPC
  • Visualization: Tableau | PowerBI
  • CLIs: bash | zsh | powershell

Technical Skills
  • Data mining / Processing (data cleaning, data processing for descriptive and predictive models, experience in ETL over multiple warehouses and sources)
  • Data Visualization (matplotlib, Altair, Jupyter Notebooks, integration into Microsoft Office)
  • Natural Language Processing (NLP) / LLMs
  • Computer Vision
  • Statistical Analysis
  • Experimental Design
  • Linux (Ubuntu, Debian, Kali Linux)
  • CI/CD Pipelines

Currently Studying

I personally love learning new things in the tech space. My interests here are not explicitly for career development, but I enjoy the art of learning how things work and rounding out my skills.

Full-Stack Web Development with React, Basic Pentesting and Cybersecurity Methodologies, Snowflake/Azure/Cloud Computing and Storage Technologies


Projects



Financial Forecasting and Backtesting with Machine Learning (Capstone Project)

This project aims to perform a survey of financial forecasting techniques with machine learning for use in stock price prediction. To reduce noise in the predictions, we decided to predict the following day’s high price from the current day’s closing price. Additionally, this also allows for ability to potentially generate alpha in actual application, despite a potential negative return day-over-day. The predictions are then run through a strategy back tester as a form of cross-validation and efficacy testing. Interestingly, success in prediction does not always correlate to success in practical trading.

Project Report




Value Stock Screener (Milestone I Project)

This was an early project in the program, designed to create a data pipeline that scraped for the current S&P 500 ticker list, pulled data from Quandl (now Nasdaq) and Yahoo! Finance, and then through modular code filters for value stocks and returns a dataframe report of possible options and corresponding investing metrics.

1-Year returns from the stocks listed in the report were as follows (adj. prices as of 12/2021):

Symbol Price 09-25-2020 Price 09-24-2021 Return (Adj.)
MET $36.26 ($34.69 adj) $61.36 ($60.90 adj) 75.6%
XRX $17.98 ($16.98 adj) $20.77 ($20.52 adj) 20.8%
HIG $35.80 ($34.36 adj) $69.40 ($69.01 adj) 100.8%
MS $47.04 ($45.61 adj) $102.91 ($102.20 adj) 124.1%
PNC $104.90 ($101.09 adj) $194.50 ($193.29 adj) 91.2%
KIM $11.22 ($10.76 adj) $21.60 ($21.44 adj) 99.3%

Given the success of the returns, I am in the process of refactoring the Jupyter Notebook scripts into OOP-based code instead of functions, and the selected securities (both these and subsequent iterations of the script outputs) are in paper trading for long-term efficacy in changing markets. Exit strategies are being researched in a back testing framework. Next steps include integration into bots for trading signals, or, regulations permitting, active trading bots.

Project Report




Proof-of-Concept RevOps Dashboard

This project was an effort to develop some revenue ops/pricing and monetization visualizations on some mock "used car sales" data from Mockeroo.

Project Repo




Microprojects Porfolio

A Jupyter notebook portfolio of non-proprietary microprojects and code snippets, including academic, self-study, and hobby projects. The purpose of these personal coding exercises is to demonstrate work with various libraries and explore alternative domain concepts.


Constantly under construction!

Portfolio Site




Disclaimer: Any financial work presented is for educational and research purposes only and is not intended to be used as an official investment recommendation. All investment/financial opinions expressed in these documents are from the personal research and experience of the project owner(s) and are intended as educational/informational material. Although best efforts are made to ensure that all information is accurate and up-to-date, occasionally unintended errors or misprints may occur in the data. There are no guarantees of financial returns based on any of these analyses. It is important to do your own analysis before making any investment based on your own personal circumstances. All financial decisions should be made with the help of a qualified financial professional.


References

    References available upon request.