Yuqing Lin

Based in · San Francisco Bay Area, United States

Hi there, I am Yuqing! I am an experienced data scientist, applied scientist.

I currently work at IBM, focused on automating order taking for McDonald's drive-through services through our voice-based conversational agent.

My focused fields are: Natural Language Processing, Natural Language Understanding, Anomaly Detection, Time Series Prediction, Product Analytics, and Experiment Design.

Experience

IBM / McD Tech Labs

Senior Data Scientist, Applied Scientist

IBM acquired McD Tech Labs on December, 2021. Techlabs has designed and is currently improving and scaling a robust voice-based conversational agent for use in the McDonald's drive-through services.

Key projects I have worked on including:

1. Integrated a Transformer network in Pytorch for Transcoder (a sequence-to-sequence model to translate customer natural language to Mcdonald’s order logic transcode)

2. Developed an end-to-end g2p (grapheme-to-phoneme) model

3. Incorporated the g2p model for Transcoder training to improve noise robustness

4. Automated our Dialogue Diagnostic process, which enabled a systematic examination of our extensive customer dialog corpus and facilitated data-driven enhancements to our dialog agent

5. Strengthened our data reliability by utilizing Argo for optimized pipelines and Great Expectations for automatic detection of data quality

Key Skills: Python, SQL, Natural Language Processing, Natural Language Understanding, Simulation, Product Analytics

February 2021 - Present

Freedom Financial Networks

Data Scientist

Freedom Financial is a fintech company helps consumers with debt relief solutions and unsecured consumer loans.

My work was focused on risk modeling and product analytics.

Key Skills: Python, SQL, Statistical Models, Product Analytics

January 2018 - January 2021

TIBCO Software

Analyst

It's an analytics and business intelligence platform for analysis of data by predictive and complex statistics.

My work was focused on opportunity scoring, experiment design, and data analytics.

Key Skills: Python, SQL, Statistical Models, Experiment Designs

January 2011 - April 2017

Education

University of California at Berkeley

Master of Information and Data Science
August 2017 - December 2019

Beijing Institute of Technology

Bachelor of Science
September 2009 - June 2013

Skills

Programming Languages & Tools
Programming Languages: Python, SQL

Tools: Argo, Amazon Quicksight, AWS, Git, Linux, Jupyter Notebook, Tableau, Superset, MS Code

Models: classific models: logistic regression, random forest, grandient boosting, time series, tf-idf; deep-learning models: LSTM, transformer, BERT

Expertise: Machine learning, deep learning, natural language processing, databases, data engineering

Projects

Code Examples & Projects
Chess Puzzle Scraper: This repo contains a chess puzzle scraper that can scrape chess puzzles from user defined web or local files. It includes testing codes and command line codes. https://github.com/graceyqlin/chess_puzzles

Code Examples: You can find some sample codes I wrote, including model implementation, algorithm questions, and hackerrank challenges here https://github.com/graceyqlin/code_examples

Visualization Search Engine: This project built a search engine that allowed visualization professionals to find relevant information more easily https://github.com/graceyqlin/visualization_search_engine

Interests

Outside of work, I enjoy hiking, cooking, running, and making silly jokes :)

Contacts

Feel free to say hi to me on LinkedIn or email me at grace.yq.lin@gmail.com. I am happy to chat and learn more about you!