As a part of our final project, we aim to analyze the likelihood of a Congress Bill to be passed given a set of variables.

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2F2c7oWsY4FcouBuwFgoDb%2Fde17a917-142e-483c-9081-6e7eb78b1d70.png?alt=media&token=db7bc51f-2667-449d-8a96-93afd76895a5","id":"de17a917-142e-483c-9081-6e7eb78b1d70","width":1164,"height":660,"filename":"201509044-ea9ce453-6db1-4a4d-96ce-b736c525d0d7.png","type":"image/png","caption":"","border":false}]}
```

Understanding the increasingly important process of forming regulations and laws in the country, we decided to explore the possibility in determining the likelihood of a proposed bill being passed by the House. This exercise can be useful for lobbyists, corporations in formulating political strategies, or even the general public during voting periods.

**Problem Statement:**

> To discover relevant factors that can help to predict the likelihood of a Bill being passed in the House

Dataset used in this exercise comes from [ProPublica](https://projects.propublica.org/api-docs/congress-api/)

*Member* - Details of Congresspeople in the House in each session with supporting information on leanings, gender, time served on the House

*Bills* - Details on the bills originating in the House in each session with supporting information on policy areas, sponsor, cosponsors, date of introduction, outcome of bill etc.

**Feature Vectors**

* Policy area

* Is the bill introduced by a Congressperson from the majority house?

* Number of cosponsors

* % of cosponsors from the majority house

* Is the sponsor in a leadership role?

* Political leaning of the sponsor [dwNominate](https://en.wikipedia.org/wiki/NOMINATE_(scaling_method)) Seniority of the sponsor

* % of votes against the party (sponsor)

Using the different models and optimization methods, we produced the following result:

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2F2c7oWsY4FcouBuwFgoDb%2Fe0f94ea6-4ee4-418d-aff3-833067340430.png?alt=media&token=fcca1654-b919-438a-911c-63212a2d9461","id":"e0f94ea6-4ee4-418d-aff3-833067340430","width":1162,"height":659,"filename":"201509275-12844e9b-4244-4e8e-b64b-3465baa00832.png","type":"image/png","caption":"","border":false}]}
```

### Main Findings

By observing the best model, we can draw several key insights as follows:

* Higher polarization leads to a higher success/fail rate of passing a bill

* Number of cosponsors in majority help in passing a bill

* Certain policy areas have higher likelihood of being passed (i.e. National Security, Emergency Management, etc.)

### Use cases

*Law Firms and Clients* - Consider a world where law firms are able to reach out to their clients and alert them that a bill is more likely to pass and explain how it would impact their business.

*Corporations & Lobbyists* - Our model has the ability to recreate how companies interact with Washington, helping them make decisions about investing efforts in the political process.

*Societal Impact* - Citizens will have the ability to stay informed about important bills impacting their personal lives like health and employment related bills. Political outcomes can be volatile and impact financial markets. Using our model, the prediction can promote stability.

Read more about the project [here](https://github.com/brianmulyadi/Designing-Data-Products)

Designing Data Products

안녕하세요!

Hey there, it's 2024, and this year, I've decided to tackle a new language: Korean. I've always been drawn to its awesome culture and incredible food, not to mention I've got some Korean pals I'd love to chat with in their native tongue!

Follow along as I jot down my weekly language adventures in this journal. Being a visual learner, it's my go-to for tracking progress—and who knows, maybe one day I'll pass on what I've learned to others.

Stay tuned for updates on this exciting journey!

**Week 1**

This week, we'll start with one of the most commonly used particles in forming sentences in Korean: **`은`** and **`는`**.

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2FKi3fUiCU9SNSCOun8vnj%2Fd6629f8a-b9df-420a-afe5-21ff6bda5c84.png?alt=media&token=4a6f8cda-461f-41e5-a916-e14087b2d457","id":"d6629f8a-b9df-420a-afe5-21ff6bda5c84","width":500,"height":500,"filename":"Week 1 - 1.png","type":"image/png","caption":"","border":false}]}
```

These particles are known as **topic markers**, used to indicate the topic of the sentence. The closest English equivalent is "*As for something.*"

For example, the sentence 저는 브라이언 이에요 can be translated as "As for me, I am Brian."

But why are there two versions, 은 and 는?

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2FKi3fUiCU9SNSCOun8vnj%2F7ba86464-da98-4e1f-ab48-c2ba7d01e576.png?alt=media&token=166032b0-937a-43c8-8559-06be849f8050","id":"7ba86464-da98-4e1f-ab48-c2ba7d01e576","width":500,"height":500,"filename":"Week 1 - 2.png","type":"image/png","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2FKi3fUiCU9SNSCOun8vnj%2Fa575fb61-8486-4931-8a6d-47709e613381.png?alt=media&token=6cab4677-c8d0-4913-8bc9-b3d4f4cf1173","id":"a575fb61-8486-4931-8a6d-47709e613381","width":500,"height":500,"filename":"Week 1 - 3.png","type":"image/png","caption":"","border":false}]}
```

It's because in Korean, the final consonant of the preceding word determines which particle to use. Let's take the example of the previous sentence's topic, 저. The final consonant of this word is ㅓ, which is a vowel. Therefore, you pair this topic with 는.

On the other hand, if we take the topic 오늘, which means "today," the final consonant is ㄹ, which is a consonant. For this, you would match it with 은.

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2FKi3fUiCU9SNSCOun8vnj%2F0b560ed2-9b6b-46a2-ab48-998ad372a623.png?alt=media&token=1ec27867-ccc1-4d8f-9a11-87f362390915","id":"0b560ed2-9b6b-46a2-ab48-998ad372a623","width":500,"height":500,"filename":"Week 1 - 4.png","type":"image/png","caption":"","border":false}]}
```

Initially, this may seem complicated, but as you continue learning, it becomes intuitive because matching these particles helps in easier pronunciation.

See you next week!

Learning Korean

Afterlife Garden is an interactive device to help children understand and better connect to beloved family members who have passed away.

Present in many cultures around the world, families typically have a dedicated corner in their houses to celebrate the memory of their loved ones. We explore this opportunity to re-imagine how we can introduce subtle interaction to evoke emotions from the users.

Read more about the project [here](https://github.com/brianmulyadi/Interactive-Lab-Hub/tree/Fall2022/Final%20Project)

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2FYLioiaYPaBb3qJm1Qa83%2F6e9471da-fc45-4966-80a1-cba1851e49b4.mp4?alt=media&token=09142c39-d7e0-4448-a17d-efb138137c25","id":"6e9471da-fc45-4966-80a1-cba1851e49b4","width":480,"height":854,"filename":"afterlife_minn.mp4","type":"video/mp4","caption":"Afterlife Garden reels","border":false}]}
```

**Interactions:**

1. Users will pick one of the miniature urns and place it in the slot

2. The capacitive sensor will sense the touch and trigger the lights to follow the movement

3. Using p5.js, the screen will show the picture and biography of a family member assigned to that slot

4. The family member's favorite song will also automatically be played by the device

**Toolkit used:**

* Raspberry Pi

* Capacitive sensor (Adafruit)

* ServoKit

* Neopixel lights

* Laser-cutting

### Presenting the device on Open Studio day at Cornell Tech

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2FYLioiaYPaBb3qJm1Qa83%2Fe6d9043e-4a88-4621-a3e6-a37b0b549718.jpg?alt=media&token=d1c82b28-5d42-4d0a-870b-ceaf467ac7c6","id":"e6d9043e-4a88-4621-a3e6-a37b0b549718","width":960,"height":1280,"filename":"photo_2022-12-08_11-52-30.jpg","type":"image/jpeg","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2FYLioiaYPaBb3qJm1Qa83%2Fa481acad-b6a8-43a8-8d54-340b6bbdb18d.jpg?alt=media&token=41cd50cf-6c1b-4984-92df-1b49bfe1508b","id":"a481acad-b6a8-43a8-8d54-340b6bbdb18d","width":720,"height":1280,"filename":"photo_2022-12-08_11-52-32.jpg","type":"image/jpeg","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2FYLioiaYPaBb3qJm1Qa83%2F9bbebc3a-8a4d-4d35-926c-ba1037379d23.jpg?alt=media&token=8570aad6-cf4d-4083-88b5-058df4b4284f","id":"9bbebc3a-8a4d-4d35-926c-ba1037379d23","width":606,"height":1280,"filename":"photo_2022-12-08_11-52-34.jpg","type":"image/jpeg","caption":"","border":false}]}
```

Afterlife Garden

Using Jupyter Notebook, I created a simple data pipeline that pulls from [Ergast API](https://ergast.com/mrd/) to generate a visual that evaluates Formula 1 drivers' performance.

At every race, drivers have to complete a qualifying lap on Saturday and race on the next day. Qualifying result will determine the position of the grid for the race. This analysis will serve the function to observe the relationship between the two-day results.

Final product is then visualized using the Seaborn library `sns.heatmap()`

Check out the codes at my [Github](https://github.com/brianmulyadi)

### Final result:

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2FbEZdprQJN2HY5qmZb0Za%2F4bb31854-9e68-4b5b-9139-70bb3fc9c295.jpg?alt=media&token=304512d2-020e-4e96-8852-ca9ff229355e","id":"4bb31854-9e68-4b5b-9139-70bb3fc9c295","width":1600,"height":2000,"filename":"formula-1.jpg","type":"image/jpeg","caption":"","border":false}]}
```

This visual is very useful in analyzing the performance and consistency of each driver. Using Jupyter Notebook, the analysis workflow is repeatable and easy to adjust to accommodate other variables that needed to be explored.

Data Visualization on Formula 1

If you are not familiar with the chess clock, it consists of two countdown timers to track the amount of time left for each player to make a move. Every time a player makes a move, they will press the button to pause their time and resume the opponent's time. Once your time is up, you lose the game!

The requirements for this product:

* It has to show two countdown timers

* The user can set the time before starting the timer

* Player A's timer starts when the program is run

* When Player A's button is pressed, the countdown timer for A stops and the countdown timer for B starts at the same time. The color of the boxes also has to be swapped.

* When one player's timer reaches 0, show "Game Over" message

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2FRuN2wvlytjhhUUUc86ezhuWFjb72%2Fy7kZhPzGcqtGaogVKmb0%2F52f6aa62-2066-4b86-895f-d0be9f4e5b55.png?alt=media&token=d3313f94-a4f9-4bdc-9dee-a88da8173aec","id":"52f6aa62-2066-4b86-895f-d0be9f4e5b55","width":785,"height":952,"filename":"191149513-4011da9c-8f82-4eda-838c-c1f2205f4757.png","type":"image/png","caption":"Storyboard and setup","border":false}]}
```

To extend the functionalities of the chess clock, I plan to incorporate some of the sensors available and connect them to the Raspberry Pi.

I use the [Proximity Sensor](https://www.adafruit.com/product/3595) and the [SparkFun Qwiic Button](https://www.sparkfun.com/products/15932) for this project.

Here is how it is in action:

```cv
youtube
{"url":"https://www.youtube.com/watch?v=HgYfRfWJj0Q","width":200,"height":113,"thumbnail":"https://i.ytimg.com/vi/HgYfRfWJj0Q/hqdefault.jpg"}
```

Shout out to [Panda Xu](https://github.com/30PandaX/Interactive-Lab-Hub/blob/Spring2021/Lab%202/cli_clock.py) and [Joshua Schmidt](https://github.com/jschmidtnj/Interactive-Lab-Hub/blob/Fall2021/Lab%202/colorbar_clock.py) for the inspiration at Designing Interactive Devices class.

Touchless Chess Clock ♟🕰

Product Builder

2nd Place – Product Games Case Competition

Representing Cornell University, we presented on our brainchild, MomumentDAO, which helps conservation projects to issue NFTs in order to create additional funding sources and to reach a global audience.

The annual Product Games Case Competition had 185 participants from 32 nationalities and 22 business schools.

1st Runner Up

1st Place – Leading Teams Case Competition

2nd Place

Developed the business model and market sizes in Trac3y – a blockchain-based solution to resolve trade disputes via smart contracts, achieving 2nd place in Penn Hackathon (out of 20+ universities)

Mentored 16 founders from ideation to pre-seed stage in a startup accelerator based in Southeast Asia

Cohort President

Graduated Magna Cum Laude, Teaching Assistant (2013), Dean’s List (2012-2014)

Recipient of the ASEAN Scholarship by the Singapore Ministry of Education

VP at Cornell Blockchain Club
Dean's List Summer '22, Fall '22

On March 1, 2018, IFC launched the report Driving Toward Equality: Women, Ride-Hailing, and the Sharing Economy, which explores how women and men participate in ride-hailing, particularly in emerging markets.

Driving Toward Equality: Women, Ride-Hailing, and the Sharing Economy

Kill/Bill – US Congress

New resolution this year to learn a new language! I love the Korean language and have always wanted to speak it fluently. I am a very visual learner, so as I learn the language, I will be creating and sharing visuals that have helped me to learn more effectively!

Learning Korean!

Representing Cornell University, we presented on our brainchild, MomumentDAO, which helps conservation projects to issue NFTs in order to create additional funding sources and to reach a global audience.

Monument DAO

An interactive device powered by Raspberry Pi to reimagine how we educate children of family members who have passed

A project to visualize Formula 1 drivers' performance in the 2021 season

Developed a JavaScript Boost for Arc Browser Hackathon to help users find product reviews directly on Amazon pages. The web extension was shortlisted to the top 10 products out of 500+ submissions

Youtube Reviews on Amazon

Trac3y

To explore my interest on IoT devices, I experimented with Raspberry Pi and Adafruit sensors to create a touchless chess clock.

Designing Interactive Devices

Presented as a guest speaker held in conjunction by Yale School of Management and University of Indonesia

Building Startup in Indonesia

Organized an annual Christmas Fundraising project in partnership with HOPE Worldwide to raise $20,000 to benefit underprivileged children and families in Singapore

Project Lead

Advised 3 engineers to create a mobile app for measuring the internet coverage and optimize antennas placement in Naluja, Zambia

Advisor

Facilitated a para-athlete basketball tournament in Southeast Asia

Volunteer Coordinator

LDC is a Big-4 global commodities firm trading $50 billion of agriculture, shipping, and financial products.

- Collaborated with Consensys to develop a platform for digitizing global trade processes using blockchain solutions, increasing transaction speed by 70% and decreasing error rate by 80%
- Launched and owned an internal data analytics platform to structure and analyze trading information: resulting in 80% faster decision-making process and 20% increase in profitability
- Managed cross-functional teams across six countries to purchase, process, and deliver $15 million of products annually to destinations within Asia-Pacific region
- Collaborated with engineering and analytics team to initiate a geospatial analysis of cargo ships: allowing new data-centric trading strategies and increasing the price forecast accuracy by 20%

Asia-Pacific Program Manager

- Optimized cargo unloading schedule to speed up discharging process, removing bottlenecks and maximizing production capacity by increasing the utilization rate of the port by 40%
- Negotiated sales and shipment contracts with the largest importer of consumer products in Philippines: securing a long-term deal worth $500,000 in annual profit
- Created strategic partnerships with 3 agricultural and shipping companies: increasing service level to 95% and expanding annual sales volume by 15% year-on-year

Oilseeds and Freight Trader

Ladyjek is a $2 million pre-series A ridesharing startup for women empowerment

- Founded an on-demand transportation venture to create employment opportunities for underprivileged women in Indonesia: partnering with over 3,000 drivers, achieving annualized revenue of $680,000
- Led product, engineering, and design team to build two main features, transportation and food delivery within 3 months: growing the user base by 300% month-on-month
- Scaled up operations by 50% by forming agreements with 3 e-commerce companies to be their exclusive logistical partner: adding reliable demand to increase average rides per driver by 140%

CEO and Founder

Kapita is a website to help team managers automate repeatable tasks with a no-code drag-and-drop workflow builder

Brian Mulyadi

2024

2022

2021