Hi There, I'm

Yang Ming.

AI/ML SPECIALIST WITH 8 YEARS OF EXPERIENCE

Expert in AI Agents (MCP/Agentic), LLM Fine-Tuning, Computer Vision, and Generative AI—driving real-world impact in traffic, healthcare, retail, entertainment, and sports.

Contact Me
About

About Me

With over 8 years of experience in AI and machine learning, I specialize in building intelligent systems that solve real-world problems across various industries. My core expertise includes developing AI agents using Model Context Protocol (MCP), designing agentic AI workflows, fine-tuning LLMs and ML models, advanced prompt engineering for generative AI, and implementing computer vision solutions. I’ve applied these skills in diverse fields—optimizing real-time traffic prediction and routing algorithms for smart cities, enhancing diagnostic support tools in the medical field through deep learning-based imaging analysis, and building AI-powered recommendation systems for e-commerce platforms. In the entertainment sector, I’ve developed generative AI systems for dynamic content creation, while in sports analytics, I’ve built computer vision pipelines for real-time player tracking and performance prediction. My focus is always on translating cutting-edge AI/ML research into high-impact, production-ready applications that drive measurable value.

Tech Stack

Python C++ JavaScript Bash SQL PyTorch TensorFlow Keras XGBoost LightGBM Scikit-learn GPT (OpenAI) LLaMA Claude Hugging Face Transformers Whisper Coqui XTTS RAG Piper OpenCV YOLO Segment Anything FiftyOne Pandas NumPy Redis AWS GCP AutoML Vertex AI MongoDB FastAPI Docker ONNX Streamlit LangChain Weights & Biases MCP Text-to-Image CrewAI Autogen Detectron2 MediaPipe DeepSort CNN Pose estimation

Education

Bachelor of Science in Computer Science

Nanyang Technological University, Singapore

2011 – 2015

Services

My Services

AI & ML Development

Build intelligent systems using MCP, agentic AI, fine-tuned models, and computer vision to solve real-world problems across multiple industries.

Generative AI & Prompt Engineering

Design LLM-based applications with advanced prompt engineering and fine-tuning for content generation, chat systems, and enterprise AI workflows.

Domain-Focused Applications

Deliver AI solutions in traffic, medical, e-commerce, entertainment, and sports, tailored to each industry’s unique challenges and use cases.

Data Analytics & Predictive Intelligence

Extract actionable insights and forecast trends from large datasets using machine learning, statistical modeling, and predictive algorithms.

Automation & Intelligent Agents

Develop autonomous AI agents that plan, reason, and interact across complex workflows to streamline operations and boost productivity.

Scalable AI Infrastructure

Deploy scalable, cloud-integrated AI systems using Docker, FastAPI, and cloud platforms like AWS and GCP for real-time, production-grade performance.

Experience

My Experience

  • Feb, 2024 - Present
    Senior AI/ML Developer

    Built an AI shopping assistant agent that scrapes real-time product prices from Walmart and Amazon, recommends the best deals, and suggests applicable coupons based on user-selected items. The agent automates coupon application and proceeds with optimized checkout, enabling users to save money seamlessly—similar to the experience offered by Perplexity AI’s shopping assistant.

    - Tech Innovators, San Francisco, USA
  • Aug, 2022 - Oct, 2023
    AI/ML Developer

    Developed an AI-powered routing and pricing assistant that optimizes ride assignments during peak hours by balancing rider satisfaction and driver availability. The system dynamically adjusts prices, supports advance booking, and leverages historical demand patterns to improve matching efficiency and reduce wait times.

    - Data Solutions Inc., New York, USA
  • Jun, 2021 – May, 2022
    Junior AI Developer

    Built a generative NFT creation system that uses DALL·E to produce unique images and GPT to generate descriptive, themed captions, enabling automated NFT asset generation at scale. The system supports customizable prompts and styles, streamlining the creation process for digital art marketplaces and collectors.

    - AI Startups, Boston, USA
portfolio

My Portfolio

Deep Channel

Develop and Optimized custom Data Pipeline

Python Prefect Redshift Snowflake AWS Docker SharePoint MySQL ThoughtSpot
I implemented robust error handling and data validation to ensure consistent data quality, reduced deployment time by implementing CI/CD with AWS CodeBuild, optimized transformation logic for handling large volumes of diverse data, successfully connected multiple data sources into a unified warehouse, and structured the data specifically for optimal performance with ThoughtSpot.

CYFOX Cybersecurity

Develop and Optimized custom Data Pipeline

React Electron Node.js SQLite Windows Desktop Application Multiple API integrations (EDR, XDR, SOCaaS) SharePoint MySQL ThoughtSpot
Designed and implemented a secure, high-performance Electron-based desktop application for real-time security monitoring by integrating multiple cybersecurity APIs (EDR, XDR, SOCaaS) into a unified system with standardized data processing, while ensuring process isolation and minimal latency through an efficient AI-powered event pipeline.

RIOS Intelligent Machines

RPA Machine Learning AI, Edge Computing

Metaflow Roboflow MLFlow YOLO-based models FiftyOne Encord Kubernetes Docker Custom frame handling AWS S3 MongoDB
Developed a production-grade MLOps pipeline for video processing at RIOS, integrating dataset management and annotation platforms (Voxel51, Encord), extending FiftyOne with custom robotics plugins, and designing scalable, reproducible ML workflows with Kubernetes support.

Unstructured IO

ETL for LLMs

YOLOX YOLO-NAS DocLayNet layoutparser detectron2 PyTorch pandas numpy pdfminer pdf2image pikepdf Tesseract-OCR PaddleOCR Pydantic AI LangChain OpenCV NLTK
Resolved critical technical challenges including memory optimization for large PDFs, accurate document element extraction, robust Unicode handling across formats, and overall code quality improvement by fixing 300+ bugs and reviewing 500+ GitHub PRs.

AI Image Insight Explorer
Gemini AI Image Recognition Streamlit for web application Python AI and Image Processing PyCharm IDE Streamlit Data Security
Ask The Image Gemini Pro Vision App is a breakthrough mobile application that utilizes AI to analyze and provide image insights. It offers object identification, text recognition, and scene interpretation functionalities. Users can snap a photo and receive instant, detailed information about elements within the image, making it a versatile tool for education, research, and everyday life.

Plus One Company

NLP-Powered Call Analysis Platform

Python spaCy scikit-learn BERT Keras TensorFlow Topicrank Textrank Yake Autophrase Google Cloud Google Data Studio ELI5 LIME
Engineered a cloud-native NLP system on GCP that combines unsupervised keyphrase extraction, BERT-powered call segmentation, and sentiment analysis to derive actionable insights from thousands of unstructured voice conversations—leveraging custom preprocessing, ensemble techniques, and automated retraining pipelines for continuous performance improvement across sales and support processes.
Contact

Contact Me

Email

yangming.rm.0121@gmail.com

Visit My Studio

50 Nanyang Ave, Singapore

Social

Discord: evermore121

Whatsapp: +381 63 1292013

Telegram: @Evermore121

Get In Touch