Yun Luk Liu-image

Yun Luk Liu

Hello! My name is Yun Luk Liu, and I am currently pursuing a Master's degree in Machine Learning at KTH Royal Institute of Technology. I am deeply interested in the fields of AI, Machine Learning and Data Engineering!

In my free time, I love working out at the gym, reading fantasy books, or playing the violin!

about-me-image

About me

I am a passionate and driven individual with a keen interest in leveraging technology to solve real-world problems! My educational background includes an M.Sc. in Machine Learning and a B.Sc. in Electrical Engineering. My master's thesis is about clustering of Alzheimer's disease data to identify subtypes of Alzheimer's using FDG-PET scans. Throughout my Master's degree, I have worked with several different ML projects. During my bachelor's degree, I had the opportunity to work with different projects as well, such as building a Qi receiver for wireless charging and using Machine Learning to predict the accuracy of Nanopore DNA sampling. Additionally, I have gained work experience through internships, including working as a software engineer at Goldman Sachs, working with R&D of Electric Vehicles at Scania, and working with R&D of high-voltage industrial appliances at Hitachi Energy

    Education

    Master of Science in Machine Learning

    KTH Royal Institute of Technology2022-2024

    Master's thesis: Subtyping of Alzheimer's disease using Subtype and Stage Inference(SuStaIn) with FDG-PET data. The project was done in collaboration with Karolinska Institutet. The goal was to use a SuStaIn model to find different subtypes of Alzheimer's disease based on the temporal progression of the disease. Python and Pytorch was used.

    Exchange Studies, Computer Science

    University of Texas at Austin2023

    Took courses in Neural Networks, Computer Vision, Neuroscience and Linguistics.

    Bachelor of Science in Electrical Engineering

    KTH Royal Institute of Technology2019-2022

    Bachelor's thesis: Pre-analysis of Nanopore Data for DNA Base Calling. The aim of the project was to create a Neural Network to predict the accuracy for DNA samples, sampled with nanopore DNA sequencing. The work was done in Python using Tensorflow and Scikit-learn.

    Work

    Summer Analyst

    Goldman SachsJun 2023 - Sep 2023
    • Software Engineering internship working with the Data Engineering team
    • Presented the results of the internship project in front of senior stakeholders

    Summer R&D Intern

    ScaniaMay 2022 - Jul 2022
    • Worked with testing and debugging newly implemented software in Electric Trucks
    • Held a presentation demonstrating how a new software program can be used for the development of ECUs in electric vehicles. The presentation has been shared with over hundreds of people in Scania’s R&D department to aid them in adopting the software program

    Summer R&D Intern

    Hitachi Energy (former Hitachi ABB powergrids)Jun 2021 - Aug 2021
    • Designed and built several PCBs for EMI and EMC testing purposes using KiCad
    • Ensured that the PCBs were following IEC standards for creepage and clearance distances for high power circuits

    Projects

    Swedish Voice-based search of Youtube

    Technologies used: Python, Pytorch, Huggingface, gradioNov 2023 - Dec 2023
    • Fine-tuned a Small OpenAI Whisper model on Swedish broadcasting data to teach it translation and transcription of the Swedish language
    • Created a gradio app, hosted on Huggingface Spaces, that searches Youtube based on Swedish speech input from the user and then transcribe the first video found

    Twitch Chat Analyzer

    Technologies used: Python, Spark, Kafka, Cassandra, Flask, IRC socketSep 2023 - Oct 2023
    • Created a chrome extension to fetch live twitch chat streaming data via IRC socket and represented the sentiment in the chat by showing an emoji that updates every second
    • Used Kafka for fault tolerant ingestion of data and Spark for processing of streaming data
    • Retrieved the sentiment of each message using VADER sentimental analyzer
    • Stored each message in Cassandra and created a Flask app to retrieve the latest sentiment

    Music Genre Classifier

    Technologies used: Python, PytorchMar 2023 - Apr 2023
    • Created a Music Genre Classifier which uses different audio features to predict the genre
    • Implemented a CNN using Pytorch and evaluated the performance of the model using cross entropy loss for the genre

    Viola-Jones Face Detection Algorithm

    Technologies Used: MatlabMar 2023 - Apr 2023
    • Implemented a Viola-Jones face detection algorithm using haar features and AdaBoost learning algorithm.
    • By sweeping an image with smaller windows of haar feature "boxes" the algorithm is able to find faces in different positions and angles

    Check out my photos too!

    alt
    alt
    alt
    alt
    alt
    alt
    alt
    alt
    alt
    alt
    alt
    © Copyright 2024 Tim Baker