Yose Marthin Giyay

Yose Marthin Giyay

Developer & Instructor

SaCode Community

Biography

As a Papuan (of the Mee tribe) and an aspiring AI developer, I have a passion of using AI to make a positive change to my indigenous community in Papua, Indonesia. I am now focused on documenting the Ekari language, while expanding my knowledge on computational linguistics and LLMs, and how they can be used to support and promote my indigenous language that is on the threshold of being endangered. I’m currently working on organizing a workshop to build a semantic corpus of the Ekari language to ultimately fine-tune an LLM for it. My aim is to develop a mascot that will act as a tutor and as a contact point for anything regarding my ethnic region. I am also a curriculum developer and an instructor for SaCode community, the first IT community and coding school in Papua.

Interests
  • Artificial Intelligence
  • Computational Linguistics
  • Retrieval Augmented Generation
Education
  • BSc in Physics and Computational Math, 2016-2019

    Michigan State University (govt. sponsored full-ride scholarship; funding still incomplete due to wide-reaching political turmoil)

  • S1 Matematika (BSc equivalent), 2023

    Universitas Terbuka

Experience

 
 
 
 
 
PT. Nokensoft Inovasi Teknologi Papua
Developer
January 2023 – Present Jayapura, Indonesia

Responsibilities include:

  • Developing landing pages for clients in TailwindCSS and NextJS
  • Creating forecasting models for client’s business data in Colab using pandas and sk-learn
 
 
 
 
 
SaCode Community
Instructor & Contributor
October 2021 – Present Jayapura, Indonesia

Responsibilities include:

  • Developing a curriculum for intro to HTML-CSS and Python
  • Running a class of 15-20 people
  • Creating reports for the class and sponsors
  • Organizing a language documentation workshop
  • Running an AI think tank
 
 
 
 
 
NSCL, Michigan State University
Operations Student Assistant
NSCL, Michigan State University
August 2018 – July 2019 East Lansing, Michigan

Responsibilities include:

  • Maintaining parts for the main particle accelerator and reports back
  • Creating inventory reports
  • Monitoring secondary cyclotron for errors

Projects

*
Respiratory Diseases Classification
A Streamlit web app for the CNN trained to classify pulmonary diseases with respiratory sound recordings (.wav). Hosted on HuggingFace Spaces.

Contact

Email is always open. DM me on Twitter, Instagram, or Discord.