Me

Hi. I'm Amarin Aum Jettakul

อัมรินทร์ เจตน์ฐากูร
Download Resume
About

Short Bio

I'm a data scientist and full-stack data professional passionate about building intelligent systems that connect data, people, and impact. With experience spanning analytics, machine learning, and data platform engineering, I love turning complex problems into elegant, scalable solutions.

Over the past few years, I've led projects across e-commerce, mobility, and AI innovation — from recommendation engines and explainable AI to modern data pipelines. Beyond code, I enjoy bridging technical depth with business sense and helping teams grow through data-driven decisions.

While there's no Wikipedia page about me (yet!), you can explore my work and projects below.

Education

  • Chulalongkorn University
    Jun 2018 · Aug 2019
    Master of Engineer in Computer Engineering
    Chulalongkorn University
    Master's degree with distinction
    Thesis: Combined state-of-the-art deep learning methods from that period to address multiple tasks in the BioNLP-ST challenges after their official completion, achieving top-tier performance comparable to the original challenge winners.
  • Chulalongkorn University
    Jul 2014 · May 2018
    Bachelor of Engineer in Computer Engineering
    Chulalongkorn University
    GPAX 3.62 of 4.00 – First class honors
    President of the Computer Engineering Student Council
    Thesis: Researched and developed pre-trained foundation Thai NLP models, including tokenizer, word embedder, part-of-speech tagger, and named entity recognizer, utilizing the private dataset from National Electronics and Computer Technology Center (NECTEC) in Thailand. Released open-source code and provided fine-tuning approaches to enable researchers and developers to adapt the models with their own datasets.

Work Experience

  • LINE Corporation
    Nov 2023 · Aug 2024
    Data Analyst (E-commerce)
    LINE Corporation, Thailand
    • Implemented key improvements in a search engine enhancement initiative, refining Elasticsearch queries and coordinating A/B tests, contributing to a 33% increase in CTR.
    • Initialized an explainable AI project using SHAP, managing cross-functional teams (PMO, PO, MKT) to create interpretable ML models. Identified key factors for new seller success, guided feature prioritization, and increased new-seller traffic by 20%.
    • Enhanced SQL query optimization strategies, mentoring the team in advanced techniques to reduce average query execution time by 30% and improve operational efficiency.
    • Managed an ETL migration from cron jobs to Dagster, collaborating with DE, MLE, and SRE teams, increasing pipeline reliability from 95% to 98% to meet daily SLAs.
    • Led a large-scale data scraping project, managing a team of 4 to develop a solution that processed 2 records/second—24x faster than existing methods—generating 500k potential sellers monthly with a 2% CVR.
  • Purple Ventures
    Jun 2022 · Oct 2023
    Data Scientist
    Purple Ventures, Thailand
    • Developed a hotel recommendation system using sequential transformers, fine-tuned on historical user interaction data, which contributed to a 20% increase in daily bookings.
    • Implemented dynamic pricing for ride hailing, using an hourly demand-supply prediction model based on hexagonal indexing, using temporal, geospatial, and weather features of historical data, supporting 20% higher profit margins from base fares.
    • Led DS, MLE, and Operations teams in developing automated QC application using fine-tuned CV model (95% accuracy on 5000 manually-labeled images), implemented on GCP via Terraform, reducing manual work by 75% and increasing operational efficiency.
    • Managed cross-functional teams (SE, DS, DE, PO) to develop semantic search using fine-tuned transformer model on query-document relevance data, resulting in 4% CTR increase and 5% revenue growth.
  • LINE Corporation
    Jul 2021 · Jun 2022
    Data Analyst
    LINE Corporation, Thailand
    • Established 30+ ETL pipelines and designed dashboards, using Dagster orchestration, resulting in streamlined daily operations for business users and a 50% reduction in manual query workloads.
    • Created a machine learning model for customer segmentation using embedding clusters, resulting in a 20% increase in targeted campaign ROI and a 15% boost in customer retention through improved marketing strategies.
  • Sertis
    Feb 2020 · Jun 2021
    Senior Data Analyst
    Sertis, Thailand
    • Conducted product cannibalization analysis, using regression techniques, to optimize product selection for 7-11 stores and a 10% increase in individual store sales margins.
    • Contributed significantly to a data warehouse migration to GCP, helping implement Airflow orchestration and optimize ETL processes. Collaborated with DE teams to migrate 100TB and increase data processing capacity by 20%, from 1TB daily.
  • OxygenAI
    Aug 2019 · Jan 2020
    Machine Learning Engineer
    OxygenAI, Thailand
    • Built MLOps infrastructure for social-commerce chatbot on GCP, using Kubernetes and Airflow. Integrated with messaging platforms, supporting onboarding of 20 sellers, achieving 80% response rate, 15% conversion rate, and 50% reduction in manual chats.
    • Engineered a bot to automatically import sellers' products and inventory from their existing e-commerce platforms, eliminating 90% of the manual importing workload and streamlining the onboarding process.
  • HomePro
    Aug 2017 · Feb 2018
    Customer Behaviour Analysis (Consulting)
    HomePro, Thailand
    • Developed a machine learning model using customer purchase history to identify those most likely to attend the Homepro Expo 2018, targeting the top 2% of a pool of 5 million customers for SMS campaigns, saving 15M ฿ in marketing costs while maintaining the same expo revenue as in previous years.

Licenses & Certifications

  • Microsoft Certified: Azure AI Fundamentals
    Microsoft
    Issued Jan 2025 · Credential ID 177D5633D99A08CD
    Show credential →
  • Microsoft Certified: Azure Data Fundamentals
    Microsoft
    Issued Jan 2025 · Credential ID 408F4918CB426395
    Show credential →
  • Microsoft Certified: Azure Fundamentals
    Microsoft
    Issued Dec 2024 · Credential ID C93C5B48394405E9
    Show credential →

Publications

  • BMC Bioinformatics
    Relation extraction between bacteria and biotopes from biomedical texts with attention mechanisms and domain-specific contextual representations
    BMC Bioinformatics · Dec 3, 2019
    We demonstrated our proposed contributions to the BB and DDI tasks on BioNLP challenges. Our proposed model architecture can be used to extract rich lexical, syntactic, and semantic features that effectively boost the performance...
    Show publication →
  • JCSSE
    A Comparative Study on Various Deep Learning Techniques for Thai NLP Lexical and Syntactic Tasks on Noisy Data
    2018 15th International Joint Conference on Computer Science and Software Engineering (JCSSE) · Jul 11, 2018
    In Natural Language Processing (NLP), there are three fundamental tasks of NLP which are Tokenization being a part of a lexical level, Part-of-Speech tagging (POS) and Named-Entity-Recognition (NER) being parts of a syntactic level...
    Show publication →