YU-YU Shih
Cambridge,MA
Summary
Motivated AI Research Scientist and Engineer with 3 years of experience deploying cutting-edge AI solutions in the tech industry. Proven track record driving impactful AI computer vision applications. Specialize in AI vision, machine learning, and predictive modeling.
Overview
6
6
years of professional experience
1
1
Certification
Work History
AI Research Scientist
MIT Computer Science & Artifitial Intelligence Lab
Cambridge, MA
08.2024 - Current
- Conducted research on advanced AI Vision technologies to enhance current methods for Classification & Segmentation and multimodal representation.
- Gained exposure to cutting-edge developments in research science through participation in numerous themed conferences and seminars.
AI Engineer
Wistron Corporation
Taipei, Taiwan
08.2021 - Current
- Played a pivotal role throughout the entire lifecycle of the AI model development pipeline: requirements gathering, data annotation, feature extraction, model selection, finetuning, validation, error analysis, final release to internal users, and CI/CD.
- Developed an advanced AI Vision system, achieving an outstanding defect detection image accuracy of >99.7%, a 50% decrease in leakage, and a 20% reduction in overkill, to a 20% cost reduction in the inspection process.
- Fostered close collaboration with subject matter experts (SMEs) to iteratively enhance computer vision AI models using streamlined data
- Formulated sets of evaluation matrices tailored for decision-makers, employing a strategic and comprehensive 5-stage approach to AI system implementation
- Spearheaded the AI Platform Proof of Concept (PoC) initiative, conducting experiments on 4+ AI cloud computing platforms, including Microsoft Azure, Amazon AWS, LandingAI on LLM and AI vision applications
AI and Deep Learning Trainee
Industrial Technology Research Institute (ITRI)
Taipei, Taiwan
03.2021 - 06.2021
- Awarded for the Best Python Project - 'Shortest Path Algorithm', highlighting excellence in Python algorithm development and implementation
- Contributed to the 'Medical Dialog Analysis using Natural Language Processing' project, applying NLP techniques to enhance understanding in medical dialogues
Graduate Research Assistant
Cornell University
Ithaca, USA
09.2019 - 12.2019
- Orchestrated and facilitated 14-week behavioral economics experiments with 22 participants weekly
- Uncovered behavioral biases from experimental research data analysis in a research-intensive environment in close collaboration with Ph.D candidates
Financial Data Analyst
Nestlé
Taipei City, Taiwan
06.2018 - 12.2018
- Conducted comprehensive Profit & Loss analysis, providing pricing and trading term recommendations to the sales team based on profitability evaluation
- Utilized SAP enterprise resource planning software for recording, analyzing, and managing data
- Performed monthly analysis for 30+ categories of complex transaction data, employing Excel (Vlookup, pivot table) and PowerBI dashboard for marketing, sales, and financial insights
Education
Master in Applied Economics -
Cornell University
Ithaca, USA01.2021
Bachelor in Finance -
National Chengchi University
Taipei City, Taiwan06.2018
Skills
- Python
- Tensorflow/ Keras
- Image Recognition
- Large Language Models (LLM)
- Statistical Modeling
- MLOps
- Data Science
- Scrum/Agile
- Data-centric AI
- Human-centered AI
- AI Regulations & Ethics
Websites
Certification
- Generative AI with Large Language Models, DeepLearning.AI, Amazon Web Service, 01/01/24
- Artificial Intelligence on Microsoft Azure, Microsoft, 01/01/24
- Microsoft Azure OpenAI Service, Microsoft, 07/01/23
Languages
- English
- Mandarin
- Spanish
Patent Project
AI Model Training and Defect Detection Optimization, Taiwan, US, India Jan . 2023
- Background: The project addresses challenges in neural network training, focusing on object detection models and making AI accessible to users with limited expertise.
- Solution: Introduces a comprehensive and systematic execution process automating AI model training and defect detection optimization standard approach.
- Key Features: Low Code/No Code Integration, Automated Label Quality Evaluation, Fuzzy Judgment Region Identification, Iterative Model Optimization.
- Impact and Significance: Reduces effort and time for AI model training, making AI accessible. Addresses challenges in defect detection, contributing to enhanced model accuracy and efficiency.
Timeline
AI Research Scientist
MIT Computer Science & Artifitial Intelligence Lab
08.2024 - Current
AI Engineer
Wistron Corporation
08.2021 - Current
AI and Deep Learning Trainee
Industrial Technology Research Institute (ITRI)
03.2021 - 06.2021
Graduate Research Assistant
Cornell University
09.2019 - 12.2019
Financial Data Analyst
Nestlé
06.2018 - 12.2018
- Generative AI with Large Language Models, DeepLearning.AI, Amazon Web Service, 01/01/24
- Artificial Intelligence on Microsoft Azure, Microsoft, 01/01/24
- Microsoft Azure OpenAI Service, Microsoft, 07/01/23
Master in Applied Economics -
Cornell University
Bachelor in Finance -
National Chengchi University
Similar Profiles
BELAL ELSHENETYBELAL ELSHENETY
Tech Officer at Computer Science And Artificial Intelligence ClubTech Officer at Computer Science And Artificial Intelligence Club
BELAL ELSHENETYBELAL ELSHENETY
Tech Officer at Computer Science And Artificial Intelligence ClubTech Officer at Computer Science And Artificial Intelligence Club
Yi-Chun ChenYi-Chun Chen
Research Scholar at North Carolina State University, Computer Science, ARNAV LabResearch Scholar at North Carolina State University, Computer Science, ARNAV Lab