Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I'm Minh Khiem Pham, a data science engineer focused on building end-to-end data pipelines, ML models, and NLP systems that scale across large datasets. I enjoy turning complex data into actionable insights that help campaigns, products, and operations run smarter and faster. I'm currently pursuing a Master of Data Science at Deakin University, with a strong foundation in statistics, machine learning, and cloud automation. I love applying LLMS and modern data architectures to real-world problems and collaborating with cross-functional teams to deliver impact.…I'm Minh Khiem Pham, a data science engineer focused on building end-to-end data pipelines, ML models, and NLP systems that scale across large datasets. I enjoy turning complex data into actionable insights that help campaigns, products, and operations run smarter and faster. I'm currently pursuing a Master of Data Science at Deakin University, with a strong foundation in statistics, machine learning, and cloud automation. I love applying LLMS and modern data architectures to real-world problems and collaborating with cross-functional teams to deliver impact.

Minh Khiem Pham

Data Scientist, Full Stack Developer, AI Engineer, +2





I'm Minh Khiem Pham, a data science engineer focused on building end-to-end data pipelines, ML models, and NLP systems that scale across large datasets. I enjoy turning complex data into actionable insights that help campaigns, products, and operations run smarter and faster. I'm currently pursuing a Master of Data Science at Deakin University, with a strong foundation in statistics, machine learning, and cloud automation. I love applying LLMS and modern data architectures to real-world problems and collaborating with cross-functional teams to deliver impact.…I'm Minh Khiem Pham, a data science engineer focused on building end-to-end data pipelines, ML models, and NLP systems that scale across large datasets. I enjoy turning complex data into actionable insights that help campaigns, products, and operations run smarter and faster. I'm currently pursuing a Master of Data Science at Deakin University, with a strong foundation in statistics, machine learning, and cloud automation. I love applying LLMS and modern data architectures to real-world problems and collaborating with cross-functional teams to deliver impact.

Available to hire

I’m Minh Khiem Pham, a data science engineer focused on building end-to-end data pipelines, ML models, and NLP systems that scale across large datasets. I enjoy turning complex data into actionable insights that help campaigns, products, and operations run smarter and faster.

I’m currently pursuing a Master of Data Science at Deakin University, with a strong foundation in statistics, machine learning, and cloud automation. I love applying LLMS and modern data architectures to real-world problems and collaborating with cross-functional teams to deliver impact.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Language

Vietnamese

Fluent

English

Advanced

Work Experience

Data Science Engineer at Viettel Group

January 1, 2022 - February 1, 2024

Designed automatic ETL flows in Viettel Data Platform/Spark to generate lists of subscribers for weekly and monthly promotional campaigns on large raw data stored in HDFS, partitioned by day. Analyzed demographic attributes and users' consumption history to summarize feature importance and correlations, and built tree-based models to predict customer churn with 70-75% accuracy. Implemented an end-to-end pipeline on Microsoft SQL Server and Viettel Cloud for data synthesis, transformation, and model optimization to forecast consumer demand for Rang Dong Company products, with testing MAPE ranging from 10% to 15%. Deployed an LLM fine-tuned on 2GB of Vietnamese documents for text summarization in law and policy, leveraging OpenAI or Gemini APIs. Built a question-answering pipeline aligned with KPIs using encoder models and LLMs, including user intent classification (98% accuracy), NER, RAG, tools calling, and fine-tuning. Researched techniques like QLoRA, data parallelism, and few-shot p

Data Science Intern, Junior Data Scientist at Misa Jsc

January 1, 2021 - December 1, 2021

Constructed hybrid models based on LSTM and BERT architectures combined with LightGBM to classify users' industries based on invoice data from the Misa meInvoice application, achieving an F1 score of 85% on the test set. Combined Pandas and Matplotlib to analyze platform product sales data and generate insights/reports for the sales department, with data queried via Neo4j to provide additional information for decision-making.