Khoi Mai | Portfolio

Hey, I'm Khoi!

CS and DS @ Purdue University (boiler up!)

My experience

GitHub Activity

Recent contributions on GitHub

Loading contributions...

Less

Background

I'm a student at Purdue studying Computer Science and Data Science. My experience spans the full data lifecycle-from ETL processes, data modeling and feature engineering to building dashboards and delivering business intelligence.

This Summer, I'm joining Zotec as a Data Science Intern on the Revenue Cycle Intelligence team! I've worked as a Data Science Intern at Antsomi and OCB, where I developed predictive models, optimized data workflows, and created visualization tools that drove decision-making. Whether it's cleaning messy datasets, training models, or presenting findings, I love creating actual impacts through solving problems with data.

When I'm not working, you can find me watching sports(basketball, tennis, football, soccer, Formula One,...).

Skills & Tech

Libraries:

Scikit-learn TensorFlow PyTorch

Pyspark Pandas Numpy Technologies: Git Docker

AWS

SQL

Resume Schema

my career, normalized to 3NF

-- click a query above

# experience Work Experience

In [1]:

experience_df.loc['Zotec']

Out[1]:

Data Scientist Intern

accelerating 13-day ml retraining across 120M+ claims

In [2]:

experience_df.loc['Antsomi']

Out[2]:

Data Scientist Intern

89% accuracy prospect model on 3M+ interactions

In [3]:

experience_df.loc['Algoverse']

Out[3]:

LLM Research Fellow

vlm context bias - 4% accuracy drop on OKVQA

In [4]:

experience_df.query("company == 'OCB'")

Out[4]:

Data Scientist Intern

pharmacy scoring model ranking 35K+ prospects across 20+ provinces

In [5]:

experience_df[experience_df['role'].str.contains('NLP')]

Out[5]:

NLP Researcher

10K+ papers ingested at 450+/hour - lda topic modeling

In [6]:

experience_df.loc['Discovery Park']

Out[6]:

EV Adoption Research

geospatial ml on 200K+ permits & 1M+ POI records

# projects Projects & Hackathon

In [7]:

khoi.projects.get('diSECt')

Out[7]:

diSECt

RAG and RAG-to-SQL for SEC filings and earnings calls for S&P 500 companies

Visit

In [8]:

khoi.projects.get('Microsoft Hackathon')

Out[8]:

2nd Place

Chatbot for credit card type suggestions, demo used as foundation for production

In [9]:

khoi.projects.get('Elite Edge')

Out[9]:

Analytics Platform

NCAA analytics platform for college basketball, Big 10 and Purdue bias

Visit

Blog

In [10]:

projects_df.tail(1)

Out[10]:

DS Playground

building blocks for understanding data science

Visit