← В ленту

Скиллы

Python
R
SQL
C
C#
Bash
Pandas
Scikit-learn
PyTorch
TensorFlow
Git
.NET Core
ASP.NET Core
Azure DevOps
Docker
GCP
Jira
Confluence
Snowflake
Hadoop
AWS
Tableau
Power BI
Google Analytics

Опыт работы

Data Science Consultant
с 07.2025 - По настоящий момент |Breakthru Beverage Group
C#, .NET Core, BBGAIAgentLib, AIAgentWeb, Azure AI, Azure DevOps
● Debug C#/.NET Core services (BBGAIAgentLib, AIAgentWeb) to validate Azure AI agent threads and ensure proper serialization/deserialization of multi-turn messages across users, supporting persistent memory and reliable orchestration logic. ● Build CI pipelines using Azure DevOps to automate .NET build/test workflows and lay the groundwork for downstream deployment of user-facing features like thread search via asynchronous APIs and streaming output.
Data Scientist
10.2024 - 06.2025 |Breakthru Beverage Group
SQL, Python
● Segmented 1,000+ B2B retail accounts into six behavioral groups (e.g., high-volume loyalists, price-sensitive churn risks) to inform upsell and retention strategies, enabling targeted sales planning across the commercial team. ● Built 30+ revenue-weighted and categorical features to represent retailer behavior, applied PCA with component attribution for dimensionality reduction, and used KMeans and Agglomerative clustering (Silhouette 0.61) to define interpretable segment. ● Presented data-driven revenue uplift insights to C-suite stakeholders based on retailer segmentation results, and proposed a tiered upsell campaign roadmap to boost repeat order rates across high-value segments.
Data Scientist
05.2023 - 06.2024 |Concentrix Catalyst Korea
SQL, Adobe Experience Platform, Google Tag Manager, BigQuery
● Increased web event tracking accuracy by 11% ahead of a global e-commerce promotion by auditing and correcting tag discrepancies between implementation guides and live tracking data, enabling cleaner downstream analytics. ● Automated country-specific data retention in BigQuery using scheduled SQL queries, managing 150+ daily-ingested datasets across 30+ global entities and reducing manual data purging time by 40% in line with client compliance policy. ● Streamlined and standardized 20,000+ tracking tags across 100+ global e-commerce sites, reducing deployment errors by 78% in Adobe Experience Platform and Google Tag Manager and improving tracking consistency across platforms. ● Designed a GDPR-compliant SQL data pipeline used by 50+ employees to enforce regulatory compliance in Europe.
Data Analyst
11.2022 - 04.2023 |Gerson Lehrman Group Korea
Selenium, Snowflake, Python
● Engineered a Python pipeline to canonicalize company names in the database by resolving multilingual and alias variations, reducing manual review time by 75% and improving entity-level accuracy to 99.8%. ● Developed a Selenium-based scraping framework to collect and enrich missing company data for integration into internal platforms, streamlining downstream workflows and enabling 20+ team members to populate Snowflake with complete records.

Образование

Machine Learning and Data Science (Магистр)
По 2025
Northwestern University
Industrial Engineering and Management Sciences (Бакалавр)
По 2018
Northwestern University

Языки

АнглийскийРоднойКорейскийРодной