Efficiency Gain
Preparation effort reduced for a high-volume, public-source financial data workflow.
AI Product ManagerTaipei, Taiwan
Reduced recurring workflow effort by up to 92% and delivered an estimated USD 42K+ in annual operating impact across AI initiatives in finance and commerce.
STRATEGIC IMPACT
I ship AI systems that cut manual work and improve decision speed.
Cost Impact
Estimated annual operating impact from automating a high-volume review workflow.
Reliability
Reduction observed within a prototype evaluation set after retrieval improvements.
AI Systems Shipped
End-to-end AI workflows and prototypes delivered across commerce, insurance, and banking.
Selected Work
Production-grade AI delivered in high-stakes environments
Financial-Domain LLM Enablement
Cut recurring preparation effort by 92% and created repeatable evaluation signals.
Problem
- High-volume public financial material required a consistent, repeatable preparation process.
Approach
- Designed a repeatable workflow for organizing public-domain financial material.
- Created a consistent evaluation layer to support model and product decisions.
Outcome
- Reduced recurring preparation effort by more than 92% through workflow automation.
Project context
Approach
- Designed a repeatable workflow for organizing public-domain financial material.
Result
- Reduced recurring preparation effort by more than 92% through workflow automation.
Project context
Problem
- High-volume public financial material required a consistent, repeatable preparation process.
Additional approach
- Created a consistent evaluation layer to support model and product decisions.
LLM Evaluation
Workflow Automation
Product Research
AI-Enabled Operations at Scale
Automated a high-volume workflow with an estimated USD 42K+ in annual impact.
Problem
- Recurring review work and fragmented coordination were slowing operations.
Approach
- Designed AI-assisted workflows for recurring operational review and product consultation.
- Worked across product and engineering to move prototypes into repeatable use.
Outcome
- Reduced recurring manual effort and improved adoption of AI-assisted workflows.
Project context
Approach
- Designed AI-assisted workflows for recurring operational review and product consultation.
Result
- Reduced recurring manual effort and improved adoption of AI-assisted workflows.
Project context
Problem
- Recurring review work and fragmented coordination were slowing operations.
Additional approach
- Worked across product and engineering to move prototypes into repeatable use.
AI Product Delivery
Workflow Automation
Prompt Engineering
Compliant GenAI Systems Integration
Improved the evidence used to assess retrieval quality and AI readiness.
Problem
- Regulated knowledge workflows required dependable retrieval and clear information governance.
Approach
- Evaluated retrieval quality and AI readiness across selected enterprise use cases.
- Mapped information dependencies to support governed delivery.
Outcome
- Improved the evidence used to assess response quality and implementation readiness.
Project context
Approach
- Evaluated retrieval quality and AI readiness across selected enterprise use cases.
Result
- Improved the evidence used to assess response quality and implementation readiness.
Project context
Problem
- Regulated knowledge workflows required dependable retrieval and clear information governance.
Additional approach
- Mapped information dependencies to support governed delivery.
Retrieval Evaluation
AI Readiness
Data Governance
Product Analysis
Full-Stack AI Service Orchestration
Unified retrieval, recommendations, and service discovery into one prototype.
Problem
- Disconnected tools and inconsistent retrieval created friction in a concept experience.
Approach
- Designed a unified assistant experience across retrieval, recommendations, location-aware services, and visualization.
- Led product flow and end-to-end prototype integration.
Outcome
- Observed a 90% reduction in hallucination cases within the prototype evaluation set.
Project context
Approach
- Designed a unified assistant experience across retrieval, recommendations, location-aware services, and visualization.
Result
- Observed a 90% reduction in hallucination cases within the prototype evaluation set.
Project context
Problem
- Disconnected tools and inconsistent retrieval created friction in a concept experience.
Additional approach
- Led product flow and end-to-end prototype integration.
AI Product Design
Retrieval
Prototype Integration
GitHub Side Projects
Applied AI builds across product, finance, and automation
5
Repositories
18
Focus Areas
Primary Build
View on GitHubAI Banking App Prototype
Independent concept prototype; not an official bank product.
Built an AI-oriented banking app prototype with clear task routing and maintainable TypeScript modules.
AI AppBanking UXTypeScriptFrontend Architecture
GPT_OSS-Finetuned
GitHubA 20B-class open model fine-tuned with public financial-domain data.
Finetuned a 20B-class open model on finance corpora and structured repeatable evaluation workflows.
Fine-tuningFinancial NLPPrompt EvaluationExperiment Tracking
Financial-Mapping-Modal
GitHubAI-powered Financial transaction classifier.
Implemented a transaction mapping pipeline for cleaner downstream reporting and reconciliation.
Transaction ClassificationData MappingPythonFinance Ops
Yolo-v8-recipt-model
GitHubFinancial receipt recognition using YOLOv8.
Trained a YOLOv8 pipeline for receipt field recognition to reduce manual document handling.
Computer VisionYOLOv8OCR PipelineAutomation
Dailynews_Bot
GitHubDaily news Notification Bot.
Built a Python bot to aggregate and schedule daily news delivery with minimal manual effort.
Notification BotAutomationSchedulingPython
Experience
Four internships across banking, insurance, and commerce
Roles where I owned AI product delivery inside regulated, high-stakes teams.
Jul 2026 - Aug 2026
Digital Strategy and Planning Intern @ CTBC Bank
Digital Strategy & Planning
- LLM enablement and innovation strategy for the bank's digital transformation agenda.
Nov 2025 - May 2026
AI Transformation Product Manager Intern @ Shopee
AI Transformation
- AI product ownership for internal automation, from requirements through cross-team rollout.
Feb 2025 - Jul 2025
Information System Designer Intern @ Mercuries Life Insurance
Information Systems Design
- GenAI integration and AI roadmap design inside a regulated insurance environment.
Jul 2024 - Feb 2025
Anti-Money Laundering (AML) Project Intern @ Taishin International Commercial Bank
AML & AI Risk Prevention
- AML and KYC operations, plus AI risk-prevention outreach at the Taipei FinTech Expo.
Skills
Capabilities across AI, data, and financial product delivery
What I reach for to take an AI product from problem framing, through technical feasibility, to something shipped and measured.
AI/LLM
Agentic AI
RAG
LLM Evaluation
LoRA/QLoRA Fine-tuning
MCP Apps
Prompt & Context Engineering
Data/ML
Python
SQL
Machine Learning
Tableau
R
Product Engineering
React
n8n
Qdrant
SDLC
Agile
Languages
Mandarin (Native)
English (TOEIC 925)
Arabic (Intermediate)
INTERESTS
Life Beyond the Terminal
Personal systems for staying curious, grounded, and active.
