01. about me
Who I Am
I'm a data analyst and data engineer currently pursuing my Master of Science in Business Analytics at Boston University. My work sits at the intersection of engineering and insight — I care about building systems that actually work, not just demos.
At TSMC, I developed an LLM-based cybersecurity tool to detect unregistered SaaS applications, eliminating approximately 200 business days of manual review effort. That experience shaped how I think about AI: it should solve real operational problems at scale.
Outside of data, I've spent 8 years competing in volleyball — including captaining a team. The sport taught me how to read patterns under pressure and lead when it counts.
6+
ML Projects
3
Cloud Platforms
8 yrs
Volleyball
Currently interested in
- ▸ Time series & demand forecasting
- ▸ LLM applications in enterprise
- ▸ Cloud-native data pipelines
- ▸ Sports analytics
02. skills
Tech Stack
Languages
ML / AI
BI & Analytics
Data Engineering
Cloud & Databases
03. experience
Work History
Digital Marketing Analyst
Autism Today Foundation
Mar 2026 – Present
Remote
- ▸Maintain interactive reporting dashboards in Looker Studio, monitoring traffic, conversion, and engagement KPIs across 4 channels—delivering validated, analysis-ready data for operational decision-making.
- ▸Translate business requirements into technical specifications for agency and CMS partners, automating reporting workflows to reduce manual data entry and align outputs across cross-functional stakeholder groups.
- ▸Develop Social SEO and campaign strategy across 3 platforms (Instagram, TikTok, LinkedIn), driving a 5% increase in website traffic by analyzing user behavior signals and optimizing channel discoverability.
IT Data Security Analyst Intern
Taiwan Semiconductor Manufacturing Corporation (TSMC)
Jul 2025 – Aug 2025
Hsinchu, Taiwan
- ▸Built an automated Python pipeline using the Google Gemini API to extract and classify 8,600+ internal URLs/IPs—detecting 220 anomalous endpoints and saving ~280 business days vs. manual review.
- ▸Built Superset BI dashboards consolidating token usage, budget exposure, and follow-up priority data to monitor pipeline data quality and surface insights for stakeholder decision-making.
- ▸Translated policy requirements into technical classification strategies for global security and compliance teams, enabling scalable analytics rollout across all TSMC foundry sites.
- ▸Automated evidence extraction from vendor certificate documents using an LLM-based workflow, streamlining key data processing steps and reducing manual compliance review time for the security team.
Data & Operations Analyst
Taiwan Institute for Sustainable Energy
May 2024 – Aug 2024
Taipei, Taiwan
- ▸Consolidated multi-source registration, invoicing, and participant records into structured data trackers, replacing manual spreadsheet workflows and enabling accurate cross-functional follow-up with 50+ partners.
- ▸Maintained backend data accuracy across customer and participant records for a 15-person team, applying data quality protocols to support reliable operational reporting cycles.
- ▸Reduced inquiry resolution time from ~30 to ~10 mins across 20–30+ cases per event by implementing standardized triage workflows—streamlining ad-hoc processes into repeatable operational systems.
04. projects
Featured Work
NCAA Football Ranking System
End-to-end data pipeline for college football rankings using Airflow orchestration, MotherDuck data warehouse, and Bradley-Terry probabilistic modeling deployed on Google Cloud Run with interactive dashboards.
Recycling with Deep Learning
CNN model for automated waste classification into recyclable categories, trained to sort materials with high accuracy for sustainable waste management.
Kai-Wei Teng Pitch Analysis
Statcast-powered analysis of SF Giants pitcher Kai-Wei Teng's sinker-to-sweeper transformation (2024–2026), examining movement profiles, velocity trends, and outcome data.
Austin Bike Share Analysis
Visualizations examining membership types, weather impacts, and station-level activity patterns across Austin's bike-sharing network from 2014–2024.
Financial Anomaly Detection
Machine learning pipeline for detecting anomalous patterns in financial transaction data using unsupervised and supervised techniques for fraud and outlier identification.
Store Sales Forecasting
Kaggle competition solution forecasting daily sales across 54 stores and 33 product families. Per-family LightGBM models with recursive forecasting achieved an 11% LB score improvement (0.430 → 0.38465).
05. education
Background
Boston University, Questrom School of Business
M.S. in Business Analytics
Sep 2024 – Jan 2026
- ▸Predictive modeling, machine learning, and statistical analysis
- ▸Cloud data architecture — AWS, GCP, Snowflake
- ▸Data engineering with Airflow, dbt, and BI tooling
Baruch College, Zicklin School of Business
B.B.A. in Computer Information Systems · Minor: Economics
Aug 2019 – May 2023
- ▸Systems analysis, database design, and software development
- ▸Business intelligence and information management
- ▸Economics foundations for data-driven decision-making
06. contact
Get In Touch
I'm currently open to full-time roles and internships in data analytics, data engineering, and data science. Feel free to reach out — I'll get back to you.
Drop me an email
schlin590@gmail.com
Check out my code
github.com/shanelin0107
Connect professionally
Built with Next.js & Tailwind CSS · Deployed on Vercel