Hi, my name is Ahjeong Yeom.

I am also known as AJ. I am a data professional whose functionality crosses over between product management and analytics.

Download Resume Recent Projects

GitHub | LinkedIn | Plum

About Me

Through my work with multiple clients, I realized the importance of precision in data processing and predictive analytics when it comes to profitable decision-making. Therefore I have been mastering the art of data through mainly Python, SQL, PySpark, and many different packages to cut the noise in data sets, detect anomalies, and create models that answer the business challenges at hand. During my free time, I find joy in design, fashion, and data visualization. Paired with a unique blend of technology and business background, I am adaptable when choosing data tools and effective when communicating with both engineers and business stakeholders.

Programming Skills

  • Python
  • SQL
  • R
  • PySpark
  • Spark
  • Hadoop
  • Hive
  • Tableau
  • MongoDB
  • Neo4j
  • Linux
  • HTML

Other Skills

  • Branding
  • UX/UI Design
  • Business Operation
  • Partnership Development
  • Project Management
  • Product Management
  • Client Communication
  • Prototyping
  • Start-ups
  • Microsoft Office Suite
  • Google Analytics

Education

Sep 2021 - Expected Dec 2022

University of Chicago

Master of Science in Analytics (Data Science)

Coursework: Big Data Platforms, Data Engineering, Data Mining, Machine Learning, Statistical Analysis, Time Series Analysis, Predictive Analytics, Deep Learning & Image Recognition, MlOps, Digital Marketing

Cumulative GPA: 4.0

Sep 2019 - Dec 2019

Hanyang Cyber University

Visiting Student

Coursework: Python, SQL Programming

Cumulative GPA: 4.0

Aug 2013 - May 2017

University of Virginia

Bachelor of Science in Commerce (Finance & Marketing)

Undergraduate business program from McIntrie School of Commerce

Cumulative GPA: 3.5/4.0; Dean’s List 2013 Fall/Spring, 2014 Fall

Work Experience

Jun 2022 - Aug 2022

Affinity Solutions

NLP Data Science Intern

Built entity extraction models using BERT (natural language transformer) and machine learning libraries such as TensorFlow, Keras, and Scikit-learn to leverage 2 billion transaction data to predict 9 entities for the top 50 merchants, and the final model scored ~99% on precision for 2 million sample

Increased data quality and coverage for Affinity’s clients and reduced costs by setting up an automated data extraction via deep learning (e.g., increased data coverage by 95% on order number entity)

Implemented a custom data cleaning process on millions of transaction data via Google Cloud Platform and Big Query

Sep 2017 - May 2020

Grit&Glo

Co-founder, Product Management & Operation

Branding, UX/UI Design & Web/App Dev Digital Agency

Accelerated the company's digital transformation process from a US-based in-person digital agency to a fully virtual digital agency offering online products and services from branding identity, UX/UI design, Web/App development to consulting, allowing for market expansion into Asia Pacific region in 2018

Led overall business operations, partnership development, and communication with international clients

Managed complete product lifecycle (UX wireframing, development, and delivery) on web development projects using content management system (CMS) and ensured the output meets the current digital trends as well as clients’ needs

Implemented data collection and analytics systems, such as Google Analytics, providing insight into SEO and user engagement

May 2016 - Aug 2016

AlphaSights

Summer Associate

Conducted due diligence to qualify industry experts to arrange consultations with clients, such as private equity firms, hedge funds, and top-tier consulting firms; completed 30 advisory projects, generating $20,000 of revenue to the firm

Recent Projects

Patent Semantic Similarity Search System

Led a team of 4 to build a patent semantic similarity search system to streamline the prior art search step of the patent application. Sentence-BERT, FAISS and Streamlit were used.

Details

NLP Classification Models

Built NLP classification models to predict helpfulness of Amazon product reviews using PySpark on Hadoop cluster

Details

Used Car Price Prediction

Implemented Random Forest, Boosting, and Bagging Regressor to enhance the model performance, predicting used car prices

Details

Brand & Product Perception For Cosmetic Industry

Performed a sentimental analysis on web-scraped 20000+ tweets for 186 brands to evaluate brand perception using TextBlob and NLTK

Details

Educare - Deep Learning powered online exam proctoring

Built an online examination proctoring service incorporating existing OpenCV and YOLOv3 deep learning models in a team

GitHub

Gaepom - A Two-sided Community For Developers And Designers

Managed a team of 6 to build an end-to-end solution for developers and designers to match into projects in bootcamp

GitHub