Resume

Sabil's Resume

Table of contents

General Information

Full Name Salsabil Maulana Akbar
Title Data Scientist @ Tokopedia | NLP & IR

Education

  • 2016 - 2020
    Bachelor of Mathematics
    Bandung Institute of Technology (ITB)
    • Focus Area in Applied Mathematics (Optimization, PDE, Financial Mathametics)
    • Learnt Core Math Study (i.e Linear Algebra, Real & Complex Analysis, Multivariable Calculus)
    • Took intermediate Statistics Courses as well (Probability Theory, Mathematical Statistics)
    • Activities
      • Took International Modeling Competition on Fish Dyamics over course of time
      • Participated in Quantitative Finance (Quant) Portfolio Construction Challenge

Experience

  • 2021 (Sep) - current
    Data Scientist (Search)
    Tokopedia
  • 2020 (Nov) - 2021 (Sep)
    Data Scientist
    Pashouses

Open Source Projects

  • 2023
    ID ASR Dataset Generator
    • A construction of v0 Indonesia language ASR-based (Automatic Subtitle Recognition) Dataset on News Channel videos
    • Created end-to-end ETL Pipeline (will be dockerized)
    • Used langdetect module for automated subtitle cleansing (inspired from C4 data construction).
    • Keywords: ETL Pipeline, Text Processing, NLP Dataset.
  • 2023
    ID NN-based Splitter

Academic Interests

  • Advancing NLP latest models to Indonesian & its local languages
    • LLM are known to perform poorly in low-resource languages, due to diminishing data volume
    • In addition, some resources of local languages in Indonesia aren't being exploited yet in latest NLP advancement
    • Hence, I'm contributing on a project (with several masterful folks from Indonesia) in carrying this mission
  • NLP Generic Toolkit Expansion
    • Motivation: Some of the most used NLP toolkits isn't available in Indonesia, yet in its local languages
    • Hence, I'd like to partake in this mission to democratize the tool & knowledge across languages
  • Low-resource NLP Research
    • There are some Low-resource Languages in Indonesia that is syntacticly transferrable w/o significant semantic degradation from Indonesian languages
    • Hence, I'm dedicated to devote on bridging NLP into low-resource languages by this approach
    • My aspiration also includes attending higher degree education in a environment such that it will enables me on driving this vision

Other Interests

  • Cracking Math Problems (at times I'm longing for a challenge for proving/disproving a Math statement)
  • Learning to code tidily and in self-explainable manner (since I'm not a CS person during college years, I have to push myself by doing self-learning)
  • Doing a truly end-to-end deployment from the scratch, by my own hands (however I may be assisted by Stackoverflow or ChatGPT)
  • Discovering Mathamatics beauty as a "language" in real-life domain (be it in DS, ML, and AI or in Social Science)