Kangli Wu
Senior DS at Meta | Harvard Data Science MS | Former Business Consultant
- Role
- Senior Data Scientist at Meta
- Location
- San Francisco, CA, US
- LinkedIn followers
- 500 followers
About Kangli Wu
Devoted senior data scientist with 9 years of experience - 6 years in DS and 3 years in business consulting. Extensive expertise in hypothesis testing and product strategies. Well-versed in Python (Scikit-Learn, Pandas, TensorFlow), R(Caret, ggplot2) and SQL for product launches and analytics, business decision-making and ML predictive modeling. Led and coached 5 DS and cross-functional teams of 30+. Established analytics principles and frameworks.Passionate to combine my business acumen with DS expertise to solve real-world challenges.
Experience
Senior Data Scientist
Oct 2023 — Present · US
Multiple LLM AI Feature Launches - Owned the lifecycles of 3 AI products from conception to launch. Created and scaled accurate, comprehensive and consistent AI feature evaluation frameworks with wide adoption across RL.• Pioneered AI feature metrics (e.g, subjectivity), composed measurement frameworks and KPI targets that boosted rigorous feature quality assessment, leading to confident early launches with 45% user satisfaction gains and 99% feature success rates.• Led cross-team standardization of the frameworks, provided guidance and established best practices adopted by 6 projects, enhancing the quality and consistency of AI feature measurement cross-organization.• Identified data gaps and architected scalable logging schemas and data pipelines to address them; Led cross-functional analytics teams of 30+ to implement my data roadmap, enabling robust AI analytics at scale.Wearable Device AI Fairness Evaluation - Invented a large-scale statistical analysis to identify demographic disparities in AI performance. Standardized the AI fairness-check framework across RL to advance technological equity.• Designed and implemented a large-scale MW-Wilcoxon rank sum statistical analysis to identify differences in a wearable device’s performance across demographics. Led product iterations based on my analysis results that reduced the number of demographic aggressors by 65%, increasing device success by 27%, user engagement by 36%, and user retention by 42%.• Deployed the end-to-end evaluation workflow from synthetic data generation to result visualization. This framework has become a mandatory organization-wide product development standard for ensuring AI fairness.
Education
Harvard T.H. Chan School of Public Health
Master's degree, Health Data Science
2018 — 2020
Hangzhou Foreign Languages School
Science
2005 — 2011
Memorial University, Newfoundland and Labrador
Bachelor of Business Administration (BBA), BBA
2013 — 2014
Beijing Foreign Studies University
Bachelor of Business Administration (BBA), Business Administration and Management, General
2011 — 2015
Skills
- Common Sense
- Knowledge Sharing
- Punctuation
- Communication
- Cooperative
- Can Do Attitude
- Volunteer Management
- Logical Thinker
- Fast Learner
- Emotional Intelligence
- Quick Study
- Love to Cook
- Leadership
- Humor
- Teaching
- Positivity
- Highly Responsible
- Calm
- Badminton
- Energy
- Mandarin
- Planning & Organizing Skills
- Creativity
- Adjustable
- Microsoft Office
- Work Effectively
- English
- Problem Solving
- Adventure
- Kindness
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.