YANISA KAEWKHAO
Ann Arbor,MI
Quote
Growth demands a temporary surrender of security.
Gail Sheehy
Overview
10
10
years of professional experience
6
6
years of post-secondary education
Work Preference
Work Type
Full TimeGig WorkWork Location
RemoteHybridImportant To Me
Career advancementWork-life balanceCompany CultureWork from home optionTeam Building / Company RetreatsPersonal development programsStock Options / Equity / Profit SharingFlexible work hoursTimeline
Data Scientist
Kasikorn Bank
01.2022 - 02.2025
Product Manager
Kasikorn Bank
07.2021 - 12.2021
Data Scientist Intern
SnapLogic
06.2020 - 08.2020
M.S. - Data Science
University of Michigan
08.2019 - 04.2021
Benchmark Specialist
Thai Oil PCL
08.2015 - 06.2019
B.S. - Chemical Engineering
Chulalongkorn University
05.2010 - 04.2014
Summary
Innovative Data Scientist and Product Manager with expertise in predictive modeling and machine learning, instrumental in boosting app-in rates and demand forecasting accuracy at Kasikorn Bank. Demonstrated proficiency in SQL, Python, and project management, with a focus on cost savings and operational efficiency. Proven ability to lead cross-functional teams and drive product enhancements in the financial sector. Committed to leveraging data insights for strategic decision-making.
Skills
Technical skills: Machine Learning, Predictive Modeling, SQL, Python, Databricks, CDSW, Statistical Analysis, Anomaly Detection
Soft skills: Data-driven Decision Making, Cross-functional Team Leadership, Project Management
Education
M.S. - Data Science
University of Michigan
Ann Arbor, Michigan 08.2019 - 04.2021
B.S. - Chemical Engineering
Chulalongkorn University
Bangkok, Thailand 05.2010 - 04.2014
Work History
Data Scientist
Kasikorn Bank
Bangkok, Thailand
01.2022 - 02.2025
- Led predictive modeling projects, resulting in doubled app-in rates for commercial loan customers by utilizing LightGBM and Optuna.
- Enhanced ATM demand forecasting accuracy by 4% through the implementation of ensemble modeling techniques on Databricks.
- Achieved a precision of over 40% on a monthly basis for detecting nominee accounts in Cambodia by analyzing mobile banking logs and geolocation.
- Uncovered signs of unemployment fraud through detailed analysis of payroll and Social Security data.
- Utilized statistical analysis and data visualization to address high turnover rates among key positions, such as data scientists, and business analysts.
Product Manager
Kasikorn Bank
Bangkok, Thailand
07.2021 - 12.2021
- Increased user engagement on the social banking platform by optimizing SQL and BI tools in collaboration with the marketing and UX teams.
- Spearheaded the development of automation systems for improved reporting efficiency, resulting in streamlined product data management.
Data Scientist Intern
SnapLogic
Bangkok, Thailand
06.2020 - 08.2020
- Identified cost-effective transportation modes through the large-scale logistic data, using Snaplogic and Tableau.
- Optimized logistic costs by performing vehicle classification analysis with AutoML H2O Distributed Random Forest.
Benchmark Specialist
Thai Oil PCL
Bangkok, Thailand
08.2015 - 06.2019
- Founded AI club to foster innovation across departments, integrating design thinking into corporate processes.
- Developed regression models, combined with fundamental sentiment analysis, to predict benzene's market prices from multiple sources.
- Executed biennial refinery benchmarking studies with an energy consultant firm, in which the results were used to reduce working hours by 90 hours per year (equaling US$37,000 per year).
- Implemented a benchmark monitoring system and initiated forecasting tools for predictive analysis, resulting in cost minimization of $7,500 per year.
PROJECTS
An Information Retrieval-based Chatbot [Python]
- Implemented a chatbot from the Reddit conversation pairs dataset using a relevant function-based NLP ranking library, rank_bm25, resulting in a 110% improvement from the random sample baseline.
- Developed a new ranking function from BM25+ and modified TF-IDF for coronavirus-related queries to match against documents containing those answers or related information, resulting in a 2,135% improvement from the NDCG@10 baseline.
Personalized Recommendation on Sephora [Python] http://bit.ly/3cxDkUC
- Built a recommender system using Neural Collaborative Filtering, a deep learning method, from more than 300K reviews on skincare products scraped from the Sephora website, resulting in a 20% improvement over the SVD baseline.
- Developed a recommendation system for onboarding users through KModes categorical clustering technique.
Socioeconomic and Racial Bias in Predictive Policing [Python]
- Forecasted crime trends using collaborative filtering, SVD, and ARIMA time-series models on Chicago's crime data from 2001-2020.
- Utilized negative binomial regression to analyze bias in crime rates across different races.
Social Media Database
- Designed a database schema and implemented SQL scripts to store and display user information for the social media platform.
- Developed Java application to execute SQL queries for enhancing friend suggestion algorithm.
Languages
English
Advanced (C1)
C1
Thai
Proficient (C2)
C2
Affiliations
- Innovation
- Psychology
- Engineering
- Travel
- Music
- Muay Thai
- Dance
- Scuba Diving
Websites
Similar Profiles
Chea SotheanyChea Sotheany
Loan Officer at Kasikorn BankLoan Officer at Kasikorn Bank
Doan Tung LamDoan Tung Lam
Manager, Multi-Corporate Banking at Kasikorn BankManager, Multi-Corporate Banking at Kasikorn Bank
Sappaya SangsirithornSappaya Sangsirithorn
Head of Asset Quality Process Development at Kasikorn Bank PCL.Head of Asset Quality Process Development at Kasikorn Bank PCL.
Jonatan ContrerasJonatan Contreras
Student Intern at General MotorsStudent Intern at General Motors
Terrance QuinnTerrance Quinn
Cook 1 at University of MichiganCook 1 at University of Michigan