Data Scientist, Machine Learning, Security Analyst

Sofia V. Kan

************@*****.*** • 612-***-****

linkedin.com/in/sofia-kan • github.com/sapho237• Saint Paul, MN

Qualifications Summary

Enthusiastic Data Scientist experienced in predictive modeling, ETL pipelines, and cloud-based analytics using Python.

Skilled in resolving data conflicts, identifying discrepancies, and ensuring data integrity through strong analytical and problem-solving abilities. Proficient in NLP, ML, and deploying scalable solutions on Azure, AWS, and Docker. Adept at risk analysis and data flow automation in Agile environment. Recognized for effectively documenting streamlined processes and supporting enterprise systems.

Key Skills

•Data Engineering & ETL Pipelines

•Machine Learning, Predictive Modeling

•Business Data Analysis & Reporting

•Database Management & SQL

•Natural Language Processing & LLMs

•Cloud Solutions & Containerization

•Risk & Compliance Management

•Data Governance & Security

•Agile Development & DevOps Practices

Experience

Data Science Intern, MapLarge Inc., Remote February 2025 – Present

●Built and optimized predictive models and analytical pipelines for Geospatial Data Analysis, Computer Vision, Natural Language processing, and Time Series Forecasting using Python (Pandas, NumPy, Scikit-learn).

●Leveraged LLMs (GPT-4, GPT-3.5, Llama) for Prompt Routing and Comparative LLM analysis via API integrations; referenced architectures and benchmarks from Papers with Code and GitHub.

●Developed notebooks in a Jupyter-based environment, managed custom kernels, and streamlined ML workflows via Jira.

●Deployed scalable solutions using Azure and Docker; collaborated on statistical modeling and ETL processes with Data Scientists. Managed project codebases and documentation on GitHub.

Research Assistant (Data Science), University of Minnesota, Minneapolis, MN May 2024 – February 2025

●Researched privacy-focused approaches for training LLMs with emphasis on Optimal Homomorphic Encryption and Reinforcement Learning from Human Feedback (RLHF) in the Distributed Machine Learning Systems Lab.

●Utilized Hugging Face and LangChain benchmarks in Python to enhance encryption, training accuracy, and AI security.

●Built distributed ML models with privacy protocols, automated workflows via GitHub Actions & PowerShell, and documented encryption benchmarks for seminar presentations.

Identity & Access Management Student Worker, University of Minnesota, Minneapolis, MN February 2024 – Present

●Administered 200,000+ users across LDAP, Active Directory, Google, AWS, and VPN, ensuring compliance with security policies. Performed information security monitoring, incident response, vulnerability management, and assisted in penetration testing.

●Triaged security tickets, investigated anomalies, and mitigated access control risks following SOPs and Agile SAFe practices.

●Aligned workflows of IAM cross-functional teams with business goals and ensured on-time delivery of solutions.

●Automated IAM workflows and incident handling, reducing manual effort by 35%. Created test automation and improved documentation using Microsoft Office and Excel, increasing search efficiency by 30%.

Web Programming Intern, LinkUp Inc., Minneapolis, MN May – November 2024

●Engineered PHP, RegEx, and CSS scripts to extract and style job data from career websites, enhancing data consistency and server performance in a Data Engineers team. Refactored legacy code and followed SDLC principles to enhance system reliability and maintainability to meet best business goals.

●Developed data pipelines to process insights from 100,000+ companies using advanced web scraping and error handling.

●Created 2,500+ automation scripts and used SQL Server and Python to streamline backend workflows, reduce data processing time, and improve the efficiency of API interactions.

Education & Related Coursework

Bachelor of Arts in Computer Science; Dean’s List University of Minnesota, Minneapolis, MN

Associate of Arts Saint Paul College, Saint Paul, MN

Program Design and Development (C++), Operating Systems (C), Advanced Algorithms, Data Science, and Program Development (Python), Advanced Programming Principles, Data Analysis (R), Computational Algebra (Python), Intelligent Robotics Systems (C)

Key Projects

Portfolio website

●Developed a responsive and interactive portfolio using React and CSS on codesandbox.io: https://ly3ntl.csb.app/

Technical Skills

Programming Languages: Python, PHP, C++, SQL, JavaScript, R, HTML, JSON, RegEx

Tools & Frameworks: Git, VS Code, Power BI, Google Workspace, Microsoft Office, Azure, Agile SAFe, Docker, AWS, Directory Services, Privileged Access Management, Kubernetes, Jira

Contact this candidate