Letter Genie

Inspiration

I came up with the idea for Letter Genie because I realized that many recommenders, such as professors, employers, or mentors, often don't have enough time to write a recommendation letter from scratch. It's fair, however, as a result, they may ask the person being recommended to provide a draft or even a fully-written letter. While this may seem like a time-saving solution, it can compromise the integrity of the letter and raise ethical concerns: it may not accurately represent the recommender's true opinion of the recommended individual. With LetterGenie, I aim to provide a solution that saves time for both the recommender and the recommended individual while ensuring the ethical integrity of the recommendation process.

What it does

The application generates a letter according to the recommender's criteria selected, with a primary focus on saving time for both parties while maintaining the level of personalization and authenticity that accurately represents the recommender's opinion. By doing so, the app ensures fairness and transparency in the process of writing recommendation letters overall.

Additionally, the app also checks for any potential biases in the letter to ensure fairness and impartiality in the recommendation.

How we built it

To develop Letter Genie, I carefully selected a technology stack that prioritizes simplicity and efficiency. I chose to use PyScript, a framework that enables users to create rich Python applications in the browser using HTML's interface and the power of Pyodide, WASM, and modern web technologies. Additionally, I used HTML with a minimalistic amount of libraries, and only pureCSS for styling to ensure a clean and straightforward user experience.

For the recommendation letter generation, I decided to use OpenAI's language model due to its impressive ability to produce high-quality, human-like text that accurately represents the recommender's opinion while being easily understandable by the recipient. This sets LetterGenie apart from other static letter generators available on the internet, as our app provides at the same time flexible and personalized/authentic letters, rather than just one generic template.

Overall, the technology stack I used allowed me to create a user-friendly and efficient solution for recommendation letter writing that prioritizes accuracy, simplicity, and fairness.

Challenges we ran into

Developing Letter Genie was an exciting challenge for me, but it was not without its obstacles. One of the most significant challenges I faced was choosing the OpenAI language model to generate recommendation letters that were both accurate and personalized while avoiding any potential biases. I tried all available models and stopped on gpt-3.5-turbo.

Another challenge was ensuring that the app remained user-friendly while also incorporating cutting-edge technologies, such as PyScript and Pyodide. It required careful attention to detail and extensive testing to ensure that the app was intuitive and accessible for users with varying levels of technical proficiency.

Accomplishments that we're proud of

Letter Genie has achieved several accomplishments that we're proud of. Firstly, the app provides a unique solution that saves time for both recommenders and recommended individuals while ensuring the ethical integrity of the process. Secondly, the app generates unbiased recommendation letters that provide a fair assessment of the recommended individual's skills and abilities. Thirdly, the app has a user-friendly interface that is accessible to users of all ages and technical abilities. Lastly, the app's positive impact on the recommendation process has been significant, reducing stress and saving time for users.

What we learned

While developing Letter Genie, I learned how to use OpenAPI and PyScript technologies for the first time. It was also my first experience using all of the technologies except HTML. Additionally, I learned the importance of user experience design and thorough testing and debugging.

What's next for Letter Genie

In the future, I plan to expand Letter Genie's capabilities beyond corporate recommendation letters to include templates for educational institutions. I also want to add a feature that allows recommenders to highlight extraordinary skills or achievements. Additionally, I plan to improve the app's visualization capabilities to show matches between what the recommender selects and what appears in the final letter. These improvements will enhance the app's user experience and provide a more comprehensive recommendation letter writing experience.

Built With

• html
• openai
• purecss
• pyscript