Hi, my name is

Taro ʕ•ᴥ•ʔ

(tah-roh)

Data scientist, gamer, mixed martial artist, and animal lover.

About me

Self-proclaimed nerd. Having studied Data Theory at UCLA, I’m drawn to any topic that concerns Statistics, Math, and Computer Science. Luckily, those three fields cover almost everything in the modern age!

I’m interested in a wide range of topics, from finding solutions to difficult social/economic issues to building AI models for rovers in space! On a more personal note, I enjoy coaching high school wrestling, learning BJJ, and watching anime.

While working as a Software Developer at @Home Realty Group, I’m looking towards continuing my education in Statistics and Math. I love talking to all sorts of people, so feel free to reach out and have a chat!

Oh, I also like to lift 💪
  • 525 lb Deadlift
  • 315 lb Bench
  • 465 lb Squat
  • 3 pickle jars

Experience

Software Developer - At Home Realty Group
May 2025 - Present

Leading end-to-end development of internal tools and data infrastructure powering At Home Advantage, the latest venture at @HRG.

  • Designed and deployed a data pipeline that aggregates 500+ auction properties monthly, enriches listings via third-party APIs, and recommends matches based on client preferences.
  • Built a dynamic, client-specific newsletter system that automates email generation and delivery to over 100 active clients.
  • Automated manual workflows across lead management and property curation, delivering a 300x increase in broker efficiency and significantly reducing time spent on repetitive tasks.
Machine Learning Engineer - Truth Systems
May 2023 - Feb 2024

TruthSystems develops advanced hallucination detection systems for large language models, ensuring safe and reliable AI use in the legal domain.

  • Led data strategy and structural planning as the sole data scientist, enabling a successful $50k pre-seed raise.
  • Designed and deployed a real-time fact verification system to address AI hallucinations and improve model reliability.
  • Prototyped, trained, and productionized dozens of ML models within the AWS ecosystem to accelerate development cycles.
Lead Data Scientist - SocialSmyths
Sep 2023 - Nov 2023

Led a 5-person team to build a LinkedIn analytics dashboard for SocialSmyths.

  • Developed a LLM-powered webscraping pipeline to automatically collect data (LangChain, Selenium).
  • Conducted feature importance analysis using Lasso, Ridge, and Random Forest algorithms.
  • Generated personalized strategic recommendations with OpenAI’s GPT-4.
Data Science Intern - AI Camp
May 2023 - Sep 2023

AI Camp equips high school students with advanced AI knowledge and hands-on skills, empowering them to build real-world AI products without requiring a college degree.

  • As an instructor, I:
    • Led three teams through the computer vision MLOps cycle, including data collection with SerpAPI, training image classification and object detection models (CNN, ViT, OpenCV), monitoring metrics with Weight and Biases, and deploying MVPs through HF Inference API.
  • As a data scientist, I:
    • Cut hallucination rates by 50% in a GPT-powered Q/A application by leveraging retrieval augmented generation (RAG) with LangChain.
    • Enabled LLMs to use tools such as vector databases (Chroma), calculators, Wikipedia, etc. to drastically increase our gen AI’s capabilities.
Marketing Intern - RippleMatch
Feb 2023 - May 2023

RippleMatch is the recruitment automation platform changing how Gen Z finds work. By replacing job boards with matching and automation, RippleMatch eliminates the most time-intensive parts of the recruitment process for both employers and job seekers. Leading employers such as Amazon, eBay, and Teach For America leverage RippleMatch to build diverse, high-performing teams and Gen Z job seekers across the country trust RippleMatch to launch and grow their careers.

  • Contributed to RippleMatch’s marketing campaign by:
    • Automating lead sourcing and student outreach with Selenium (i.e. creating bots to do the work for me).
    • Leveraging social media, email marketing, presentations, and peer and faculty member networking to grow brand awareness.
  • Recognized by RippleMatch’s Leadership Team as top 15% in terms of intern performance out of a class of 500+ active interns.

Projects

VALORANT Ranked Analysis
Web Scraping Cluster Analysis Predictive Modeling
VALORANT Ranked Analysis
Applied clustering, dimensionality reduction, and supervised learning to identify playstyle archetypes and predictive factors of ranked success in top-tier VALORANT players, revealing the limits of conventional performance metrics.
Reinforcement Learning: Snake Game
Reinforcement Learning Deep Learning Q-Learning
Reinforcement Learning: Snake Game
Developed Q-Learning and Deep Q-Learning agents to play Snake, analyzing how state abstraction impacts learning performance and overfitting.
Modeling Customer Behavior
Behavioral Modeling Supervised Classification Feature Engineering Deep Learning
Modeling Customer Behavior
Predicted e-commerce customer purchase journeys for Fingerhut using XGBoost, achieving a 73% F1 score on real-world behavioral data.
Attorney Search! (ASA DataFest Placer)
NLP Clustering BERT (HuggingFace) Hackathon
Attorney Search! (ASA DataFest Placer)
Developed an end-to-end pipeline using DistilBERT and K-Means clustering to classify legal queries and match clients to optimal attorneys based on demographic-aware scoring.
Analyzing Class Performance
Data Analysis Multiple Linear Regression R
Analyzing Class Performance
Used stepwise multiple linear regression to identify key socioeconomic and campus climate factors impacting GPA among UCLA undergraduates, revealing strong links between academic performance and perceived discrimination, social belonging, and parental education.
Curriculum Recommendations with TensorFlow
Deep Learning NLP Python
Curriculum Recommendations with TensorFlow
Developed a Siamese Neural Network to align global educational content with curriculum topics across 150,000+ documents, achieving high predictive performance (AUC 0.86) through multilingual text preprocessing and large-scale pairwise matching.
Predicting Car Crash Severity
Random Forests XGBoost KNN Imputation R
Predicting Car Crash Severity
Built a Random Forest model with extensive feature engineering and imputation to predict car accident severity on a U.S.-wide dataset, achieving 93.5% test accuracy and ranking top 15 in a Kaggle-based course competition.
Tracking the Opioid Epidemic
Data Analysis Linear Regression R
Tracking the Opioid Epidemic
Modeled the origin and spread of opioid use across U.S. states using drug data and predictive algorithms to inform potential interventions.

Education

2020 - 2024
B.S. in Data Theory
University of California, Los Angeles
GPA: 3.74

Extracurricular Activities:

  • 2x ACM DataFest Runner Up (2022, 2023)
  • UCLA VALORANT team (2020-2022)
  • UCLA Wrestling team (2020-2021)

Get in touch!

My inbox is always open. Whether you have a question or just want to say hi, I’ll try my best to get back to you!