-
Notifications
You must be signed in to change notification settings - Fork 0
Team Page
Welcome to the CodeBae Team!
CodeBae is an IDE/browser extension that helps you write code and finish code at blazing fast speeds. While we are still ideating the exact design, we envisage the system as an AI system that receives as input a user's working directory of code and their current typing, and outputs intelligent suggestions on the code that they might be wanting to type next.
Team Member Name | Email Address | Photo | Skills | Personal Traits | Desired Growth | Weaknesses |
---|---|---|---|---|---|---|
Joshua Lara | joshlara@stanford.edu | systems programming, some front-end programming, prototyping | Prefer to work ahead of time, likes descriptive code commenting | Integration of backend and frontend, more front-end dev experience, documentation | AI, math | |
Alex Langshur | adl@stanford.edu | AI/deep learning, backend, finance, stats | Good long-term concentration/focus, experienced with collaborative design principles | Full stack integration, documentation standards, CI/CD | front-end, sometimes impatient | |
Henry Mellsop | hmellsop@stanford.edu | AI/deep learning, backend, finance, stats, parallel systems/hardware | Prefer to start work well before deadlines | Documentation standards, large-environment programming habits, full-stack experience | Can rush through work, can be too opinionated | |
Ryan Ludwick | rludwick@stanford.edu | Back-end programming, AI/deep learning | Proactive worker, detailed | Project documentaion, process of completing project from scratch | front-end knowledge, prototyping | |
Thariq Ridha | taridha@stanford.edu | ... | Aesthetics and layout, building prototypes, front-end programming | Organized, eager to develop new skills, proponent of writing good code quality/documentation | Back-end development, Systems architecture | AI, math, not assertive sometimes |
We will be communicating with each other over iMessage and Github requests/comments. To get in touch with us, use the email addresses provided.
https://docs.google.com/spreadsheets/d/1jdNeSno7Xc0z_mKq5H-wW1WgBiH1tvwBHzcoUfGzye4/edit#gid=0
Goal: The goal of the product is to speed up the process of coding by reducing typos and suggesting code completions. The machine learning model predicts likely completions and displays pop up suggestions while a user is typing. We want to learn common words and phrases so users don’t have to memorize syntax or other variable names.
Users: People who wish to increase their coding speed and avoid mistakes. Potential users are very large and include anyone that codes, from students to software engineers at companies. Specifically includes people who use code editors like vscode.
Similar products: This idea is not new with prominent services such as Kite and Tabnine. They scan the world’s code and train on open source code in order to offer suggestions while users type. They allow for an optional private code mode in which the model looks at code you write to learn on an individual level.
Customizable need: Offer more customization to predict off of a user’s previously written code, including in other repositories. Custom considerations include how many words in advance does the model predict and how many options does it give.