AWS Textract document processing diagram

document processor

AWS offers various machine learning tools that can be used alongside its core services. In an effort to learn more about AWS Lambda and S3, I implemented an intelligent document processing workflow.

To transform an uploaded image into text, Amazon Textract uses optical character recognition to analyze handwritten and structured writing. Then, Amazon Comprehend uses natural language processing to explain the structure of the document by categorizing words and phrases into pre-existing categories called entities.

plot animation

In my demonstration, I visualized a Wiener process to model a geometric growth projection. This stochastic process describes a nondeterministic trajectory, and by including a plot for standard compound growth, I highlighted the randomness associated with the path.

I used the animation module from Python's Matplotlib library to illustrate the trends of time-series data. The animation synchronizes a function with the frames of the display, where after each time increment, a new image is rendered.

react game

I am developing an interactive web-browser-based game that quizzes users on memorization and quick thinking. I started this project to build my familiarity with React and other JavaScript development tools like Node and Webpack.

These projects have allowed me to apply core concepts in React, including components, effects, and JSX. Holistically, I have gained valuable experience with functional programming, and I have also learned a lot about the Virtual DOM and how to interact with DOM elements.

social media

Here is a demo from the social media site that I created using Django, an open-source Python web framework. I set up a local development server, and I programmed APIs that handle client requests to make new posts and interact with other users.

This project required me to define data models for the post, follower, and user information that I collected. I created intuitive one-to-many and many-to-many relationships for logical storage and access. I used a SQLite database, which allows for lightweight and local data storage.