Eddie Lin

Eddie Lin

About

Detail

AI Product Leader | Ex-Meta | Industry Fellow @ Harvard Business School l Adjunct Professor @ NYU | Turning AI & Data into Measurable Business Value
United States
Contact Eddie regarding: 
work
Full-time jobs

Timeline

work
Job
school
Education
folder
Project
auto_stories
Publication

Résumé

Jobs verified_user 0% verified
  • Harvard Business School AI Institute
    Industry Fellow
    Harvard Business School AI Institute
    Dec 2025 - Current (8 months)
    As an Industry Fellow at the Digital Data Design Institute at Harvard (D³), I leverage my AI product leadership to help advance applied AI and digital innovation. I advise research labs, translate industry needs into actionable insights, and partner with faculty and practitioners to accelerate real-world AI adoption and impact.
  • NYU School of Professional Studies
    Data Science Adjunct Professor
    NYU School of Professional Studies
    Sep 2025 - Current (11 months)
  • New York University
    Adjunct Professor - Data Science & Business Intelligence
    New York University
    Sep 2025 - Current (11 months)
  • Korn Ferry
    Director of Data Science & AI, Product Development
    Korn Ferry
    Sep 2025 - Current (11 months)
    Lead the strategy and development of data science and AI solutions that advance people analytics, talent development, and skills assessment. Partner with business and technology leaders to drive data-informed decision-making and help organizations to bridge the gap between AI technology and workforce adoption for digital transformation.
  • Elevation Intelligence
    Founder & Chief Empowerment Officer
    Elevation Intelligence
    Sep 2025 - Current (11 months)
    𝗪𝗼𝗿𝗸𝗳𝗼𝗿𝗰𝗲 𝗧𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻 – Designs strategies to build AI-ready, data-literate, and resilient workforces. 𝗗𝗮𝘁𝗮 𝗦𝗰𝗶𝗲𝗻𝗰𝗲 & 𝗠𝗲𝗮𝘀𝘂𝗿𝗲𝗺𝗲𝗻𝘁 – Develops analytics frameworks and measurement systems to link AI initiatives to business outcomes. 𝗢𝗿𝗴𝗮𝗻𝗶𝘇𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝗪𝗲𝗹𝗹𝗻𝗲𝘀𝘀 – Embeds human-centered design and well-being metrics into transformation programs. 𝗖𝗮𝗽𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗕𝘂𝗶𝗹𝗱𝗶𝗻𝗴 & 𝗨𝗽𝘀𝗸𝗶𝗹𝗹𝗶𝗻𝗴 – Delivers customized learning programs on AI literacy, data science, and digital transformation. 𝗔𝗜 𝗣𝗿𝗼𝗷𝗲𝗰𝘁 𝗦𝗽𝗿𝗶𝗻𝘁𝘀 & 𝗣𝗿𝗼𝘁𝗼𝘁𝘆𝗽𝗶𝗻𝗴 – Facilitates cross-functional workshops to co-create and measure AI solutions. 𝗔𝗱𝘃𝗶𝘀𝗼𝗿𝘆 & 𝗧𝗵𝗼𝘂𝗴𝗵𝘁 𝗟𝗲𝗮𝗱
  • Correlation One
    Lead Instructor - Enterprise Data Fluency
    Correlation One
    Oct 2023 - Current (2 years 10 months)
  • Correlation One
    Lead Instructor - Enterprise Data Fluency & AI Literacy
    Correlation One
    Oct 2023 - Mar 2026 (2 years 6 months)
  • Harvard Business School
    Guest Lecturer - Design Thinking in Data Science
    Harvard Business School
    Mar 2023 - Mar 2024 (1 year 1 month)
  • Meta
    Staff Data Scientist - Quality Analytics & Insights
    Meta
    Feb 2023 - Sep 2025 (2 years 8 months)
  • IEEE
    Chair of ICICLE - Workforce Development
    IEEE
    Jan 2023 - Dec 2023 (1 year)
  • The City University of New York
    Data Science Instructor
    The City University of New York
    Sep 2021 - Dec 2024 (3 years 4 months)
  • Meta
    Sr. Data Scientist - Human Reviewers Learning
    Meta
    Jun 2020 - Feb 2023 (2 years 9 months)
    Lead AI-driven analytics initiatives that improve workforce performance, learning, and decision quality across Meta’s global operations. Combine AI/LLM, data science, machine learning, and UX research to design scalable systems that augment human capability and transform how organizations learn and adapt.
  • Springboard
    Machine Learning Engineer
    Springboard
    Dec 2019 - Apr 2020 (5 months)
    Build, deploy, teach machine learning/deep learning models
  • The City University of New York
    Data Analytics Research Fellow, Office of Research, Evaluation, & Program Support (REPS)
    The City University of New York
    Sep 2019 - Apr 2020 (8 months)
    • Conducted statistical & quantitative analysis for program evaluation & reporting • Streamlined & automated reports through interactive data visualizations • Designed assessment plans using machine learning, propensity score matching, time series analysis
  • Teachers College Columbia University
    Course Instructor
    Teachers College Columbia University
    Jul 2019 - Sep 2019 (3 months)
    • Led graduate students to design & optimize data-driven & evidence-based EdTech products using research design, cognitive intervention, data mining & learning analytics, assessment planning -- MSTU4083: Instructional Design of Educational Technology -- MSTU4052: Computers, Problem Solving, and Cooperative Learning
  • Columbia Engineering
    Researcher, Learning Analytics
    Columbia Engineering
    Jun 2019 - Aug 2019 (3 months)
    • Developed a survival analysis model using Kaplan-Meier estimate to predict user churn in 8 online engineering classes and identified user persistence rate in relation to enrollment time • Engineered important user behavioral features using random forest, regression with regularization, PCA, K-means as evidence for intervention design (A/B testing) • Predicted user learning outcomes (classification & prediction) with decision tree, random forest, XGBoost, kNN, SVM, logistic regression, linear regression • Led training workshops: supervised & unsupervised ML, survival analysis, (quasi-) experimental design, methodological triangulation, data confidentiality
  • Columbia Engineering
    Business Analytics Associate Course Manager
    Columbia Engineering
    Jan 2019 - Aug 2019 (8 months)
    4 Courses in this Business Analytics MicroMasters (1)Analytics in Python (2)Data, Models, and Decision in Business Analytics (3)Demand and Supply Analytics (4)Marketing Analytics
  • McGraw Hill
    Data Scientist Intern
    McGraw Hill
    Sep 2018 - Aug 2019 (1 year)
    • Conducted research on ALEKS artificial intelligent assessment & learning systems • Built logistic regression, random forest, C5.0 models using repeated cross-validation and upsampling for imbalanced data, achieved bootstrapped 68%/60% accuracy in predicting users who will fail/meet expected learning goals • Maintained Python & R code integrity in research projects, reproduced analysis results using Python & R interchangeably • Collaborated with senior data scientists to prepare & host workshops for junior data analysts at academic conferences, topics included data analysis & basic statistics with Python, machine learning, causal inferencing, deep learning
  • Columbia University
    IBM Wood Doctoral Research Fellow
    Columbia University
    Sep 2016 - Sep 2020 (4 years 1 month)
    Projects link: https://github.com/eddiecylin/data-analytics • Identified high- & low-performing students' characteristics in online tutoring system using K-means, built random forest, kNN, SVM model & achieve 92% accuracy of predicting outcome of a question attempt • Developed LDA, text mining, sentiment analyses based on students' course study notes, identified the variables that affected students' note posting behaviors with ad-hoc metrics • Designed a social networks analysis using different centrality measures to inspect student collaboration, isolated students, opinion leaders in the same class cohort • Measured levels of user interaction in 28 online class based on user forum posts & replies through exploratory data analysis & data vi
Education verified_user 0% verified
  • Columbia University
    Master of Science(M.S), Learning Analytics
    Columbia University
    Core data analytics skills: Machine Learning & Statistical Modeling • Deep Learning & Neural Networks • Clustering & Predictive Analysis • Time Series Analysis • Social Network Analysis • Natural Language Processing • Physiological Data Mining • Data Visualization
  • IBM
    Doctor of Education - EdD, Instructional Technology & Media
    IBM
    • Areas of research: - Applications of Artificial Intelligence in Cognitive Solutions - Machine Learning, Learning Analytics, Learning Sciences - Human-Computer, Human-Robot, Human-AI Interactions - Experimental Design (A/B testing) & Causal Inference in Statistics - Future of Work & Learning, Data Democracy, Algorithmic Transparency • IBM Ben & Grace Wood Researcher Fellowship (Full Doctoral Scholarships Awarded by Columbia)
  • IBM
    Generative AI for Data Scientists Certificate, Master of Science(M.S), Learning Analytics
    IBM
    Core data analytics skills: Machine Learning & Statistical Modeling • Deep Learning & Neural Networks • Clustering & Predictive Analysis • Time Series Analysis • Social Network Analysis • Natural Language Processing • Physiological Data Mining • Data Visualization
  • The Wharton School
    AI For Business
    The Wharton School
  • Columbia University
    Doctor of Education - EdD, Instructional Technology & Media
    Columbia University
    • Areas of research: - Applications of Artificial Intelligence in Cognitive Solutions - Machine Learning, Learning Analytics, Learning Sciences - Human-Computer, Human-Robot, Human-AI Interactions - Experimental Design (A/B testing) & Causal Inference in Statistics - Future of Work & Learning, Data Democracy, Algorithmic Transparency • IBM Ben & Grace Wood Researcher Fellowship (Full Doctoral Scholarships Awarded by Columbia)
Projects (professional or personal) verified_user 0% verified
  • C
    Coursera Forum Post Analysis: Investigating User Interaction across 28 MOOCs
    Aug 2018
    Project Summary : • The number of users in a course correlates positively with the number of forum posts with some exception across 28 MOOCs • Using an ad hoc user interaction index which considers both the frequency counts of forum posts and time lapse before post replies, it is found there is a significant difference in user interaction between courses • Students account for the majority of forum posts across the 28 courses while TAs and course instructor also play a role in forum posts in some courses • Replies to others' posts come from other various users rather than a few users, suggesting a “spread out” fashion in forum post replies. This also points out a healthy user interaction via forum post
  • N
    NLP/Text Mining: Analyzing Students' Note-Taking Behavior in A Graduate Class
    Jul 2018
    Project Summary : • The length of student notes tended to vary when based on different weekly taught topic • Students tended to write more in their notes when readings are journal articles or blog posts in this project • The submission time does affect note length and note sentiment. Note sentiment and length also correlate differently at different times during the day.
  • U
    User behavior analysis in ASSISTments online tutor system
    Jul 2018
    Project Summary : • Discovered high/low- performance users are distinguished by few/more number of question attempts & hints use • No significant difference in response time between high/low-performance users • Number of question attempts & hints use are the 2 most important features to distinguish the 2 types of users • Using 4 selected features(number of attempts, number of used hints, response time, total time spent on a question), random forest and kNN models in this project can predict nearly 92% of outcome when a user attempts a question
  • O
    Observe change in learning motivation & grouping with K-means
    Jan 2018 - Dec 2018 (1 year)
    Project Summary : • used students’ self-report motivation score, their responses about their interests, and preference about the course they participated to cluster students • discovered high/ low motivation cluster, as well as, another cluster that is somewhere in between • observed the variation in student motivation and interest throughout the course
  • P
    Predict the college course dropouts with decision trees
    Jan 2018 - Dec 2018 (1 year)
    Project Summary : • used multiple tree models and achieved 70% (specificity)student dropout prediction • used boosting, rules and trees to attempt better model performance • applied random forests and discovered features that yield highest importance(number of courses taken) in predicting drop-out
  • D
    Do students want to work with their best friends? (with social networks analysis)
    Jan 2018 - Dec 2018 (1 year)
    Project Summary : • discovered there is no single the best friend for the whole class, but there are best friends for students divided by gender • found that student do not prefer to work with their good friends(colleagues can seldom be good friends) • found socially isolated student in the class that may require course instructor's attention and assistance
  • U
    Using data analytics to predict drop-out students in MOOC & increase student retention
    Jan 2016 - Dec 2016 (1 year)
    Project Summary : • Applied statistical & machining learning methods (k-means, kNN, logistic regression, decision trees) to analyze 20,000 students’ data from a STEM online course • Achieved 20% increase in student retention & 10% increase in students that achieved the course certificate
Awards verified_user 0% verified
  • •
    • Toastmasters International Competitive Communicator (CC)
Publications verified_user 0% verified
  • J
    [Journal Article in Press] Learning for Duty or Enjoyment: Two Paths to Fulfillment in Learning for Taiwanese High Schoo
    Journal of Research in Education Sciences
    Jan 2018
    Abstract: Based on Self-determination Theory (SDT), many past studies have shown intrinsic motivation to be the key to engaging students in their academic work, achieving better school performance, and acquiring psychological well-being. In contrast, an increasing body of research focusing on Eastern Asian Confucian societies, including Taiwan, has shed light on the salience of extrinsic motivation centered around role obligation and duty fulfillment. It is found that such type of motivation has a high value of encouraging students’ effort making, enhancing their academic performance, and instilling their psychological fulfillment. To help settle this theoretical dissonance, this research constructed and validated the two learning paths, on
  • E
    [Journal Article] Developing communication strategies for mitigating actions against global warming: Linking framing and
    Environmental Communication SSCI
    Oct 2016
    Abstract: Although there is increasing public awareness of global warming, there is a gap between such perception and relevant actions to combat the problem. In order to develop effective strategies for facilitating public actions, in this paper we draw upon framing theory and a dual processing model. Based on an experiment involving 261 participants from a large public university in Taiwan, this study found that by framing global warming as a local issue, communication messages can trigger both analytic (issue relevance) and affective (negative emotions) appraisals, which, in turn, will increase people's intentions to take actions. This study provides important insights for government and environmental groups when designing communication c
  • [
    [Conference Paper] Multiple Approaches to Learning with Satisfaction: Construction and Validation of Multiple Learning A
    Oct 2015
    2015 Taiwan Psychology Association Annual Convention and the International Convention of Learning, Teaching, and Assessment
  • T
    [Conference Paper]The Moderating Role of Cognitive and Affect Heuristics in Shaping the Effect of Science Documentarie
    Taiwan Science Technologies Society STS Association Taiwan
    Jan 2012