< Daniel S. Portfolio
👋 Hi, I'm

Daniel Setiawan

I'm

R&D Engineer with 5 years of experience in hardware, engineering, and manufacturing. Possesses a strong ability to work collaboratively with teams and lead projects, with high attention to detail and analytical efficiency. Seeking to leverage professional industry experience and knowledge as a data science fellow, utilizing various ML techniques to gain business insights pivotal for company success.

About Me
Welcome to my page!

Officially, my job is to stand behind the operations where computer science and physics meet. But really, that's just a fancy way of saying I help create an opportunity for the world to advance.

I'm an alumnus of UC Santa Barbara. Intrigued by dance and design. Inspired by philosophy and writing. I seek to envision the unlikely, harvest from optimism, discover new ideas, and be surrounded by those who bring out the best in me.


Experience
  • September 2022 - March 2023

    NYC Data Science Academy

    New York, NY

    Data Science Fellow

    A highly immersive Data Science program involving data strategy and ML expertise.
    • Developed a trading analytics technique utilizing natural language processing sentiment analysis to drive alphastrategies in ~5M scraped tweets, producing returns 10-20% higher than the traditional SMA crossover benchmark.
    • Drove data-driven insights through various modeling techniques, such as: Lasso, SARIMA (timeseries), SVR, and tree-based models to predict and forecast accurate housing prices, developing effective home-flipping strategies.
    • Explored consumer behavior trends associated with AirBNB & travel in order to identify and implement innovative marketing strategies that businesses and hosts can adopt to remain competitive during economic downturns.
    • Conducted a comprehensive analysis of YouTube video analytics to uncover consumer marketing behavior, identify global buying trends, and discover which type of products is most impactful for certain regions / consumer groups.
    • Deployed 3 interactive dashboards using Python Dash & R Shiny: translated & visualized complex datasets to break down information, allowing users to predict trends and visually compare. See dashboards below (in projects)
  • September 2022 - March 2023

    Rigetti Computing

    Berkeley, CA

    R&D Engineer

    Y-Combinator 14' full-stack quantum computing company using superconducting tech.
    • Designed a python-based script utilizing SCPI to automate a switch matrix, classifying devices as pass/fail and auto-generating an HTML report, resulting in an 80% reduction in manual testing time with a 98% accuracy rate.
    • Programmed an automated a python script that pinpointed device characteristics (eg. resonant frequencies and reflection peaks), and generates a detailed characterization summary, delivering a 90% reduction in analysis time.
    • Developed and deployed a sophisticated spreadsheet-based automation tool for contingency planning, increasing visibility for wiring schematics, automatically generating bills of materials, and tracking moving components.
    • Led deployment planning for 15+ systems across 3 facilities spanning 2 countries: design comprehensive wiring schematics, audit bills of materials for procurement, and provide manufacturing hands-on support, as necessary.
    • Collaborated with 5 cross-functional teams and 2 external vendors to identify improvement areas for existing R&D components and implement design-of-experiment prototypes to enhance their performance and efficiency.
Skills
Programming Languages

Python, SQL, R, SCPI, HTML, CSS, JS

Frameworks

Tensorflow, Pytorch, Plotly Dash, R Shiny, NodeJS

Machine Learning

PCA, GLM, Random Forest, XGBoost, LSTM, CNN, Transformers (Neural Networks)

Software Tools & DBMS

R Studio, Docker, MySQL, Jupyter, Git, Firebase, GCP, Excel

Hardware Tools

AutoCAD, Solidworks, Arduino IDE, Geany (Rasberry Pi)

My Portfolio
Developing the winning trading strategy

Neural Networks, CNN, LSTM, BERT

Are we able to use twitter sentiments to generate a
trading strategy scoring higher than the benchmark?

The ultimate guide to maximizing profits

ML, Random Forest, XGBoost, Timeseries, SVR, Lasso

When it comes to purchasing a home, it's crucial to
know the where, when, and why.


How to thrive during a recession

R, Shiny, EDA

The best way to learn how to thrive during uncertainty
is to analyze a company who started during uncertainty.


Cracking the code: What should I sell?

Python, Web Scraping, EDA

How can we use YouTube data to know which products
consumers are more likely to buy, and when?

Developing a winning strategy using NLP Sentiment Analysis

With all the hype of generative AI, it's no surprise that AI is starting to play a major role outside the tech sector.

In this project, I'll be exploring how to use the techniques and methodology behind AI and transformers to create a trading strategy that often outperforms the traditional strategy.

  • Languages: Python, R, HTML, CSS, JS
  • Frameworks: Pytorch, Keras, Plotly Dash
  • Models: RF, RNN, LSTM, BERT
  • Information: Dashboard Slidedeck Blog

Maximizing Home Flipping Profits using ML Techniques

Though it may be seen as a lucritive business, over 28% of home-flippers end up losing money.

In this project, we'll be demonstrating various machine learning techniques that can help us make informed decisions about which properties to purchase and which improvements generate the highest return on investment (ROI).

  • Languages: Python, R, HTML, CSS
  • Frameworks: Shiny, Scikit Learn, Pandas
  • Models: Lasso, RF, SVR, XGBoost, Stacked, Timeseries
  • Information: Dashboard Slidedeck Blog

How to thrive during a recession

Whether you're a business or a homeowner, it's crucial to understand the short-term rental market to capture trends.

In this project, I'll be exploring how to stay ahead of the recession, with data from a company who started during a recession

  • Languages: R, HTML, CSS
  • Frameworks: Shiny, Scikit Learn, Pandas
  • Models: Lasso, RF, SVR, XGBoost, Stacked
  • Information: Dashboard Slidedeck Blog
11
Datasets
10
Countries
8
Months
400K
Observations

Cracking the code: What to sell?

As a society, we're emotionally attached to YouTube and have allowed it to play a significant role in our lives

In this project, I'll be exploring how we can take advantage of this emotional attachment by using it to find optimal products and times to sell to consumers.

  • Languages: Python, R, HTML, CSS
  • Frameworks: Shiny, Scikit Learn, Pandas
  • Models: Lasso, RF, SVR, XGBoost, Stacked
  • Information: Slidedeck Blog