Google Codey Applications

AutoML Jupyter

Many ML researchers are unhappy with their development process. Coding from scratch is laborious since the process for developing and testing new models is largely the same each time but no-code and low-code platforms do not provide enough granularity to tweak models, loss functions, and training processes. Most ML researchers experiment in Jupyter notebooks. They are quick, composable, and easy to present.However, even with the help of LLMs: - Copy-pasting code between web-interfaces and notebooks is slow - Errors in generated code are difficult to detect and fix - Writing the appropriate prompt to generate correct boilerplate code is still repetitive Our solution takes existing data and a natural language prompt and uses it to build a model that is compatible with the shape and types of the data. It also uses recursive API calls to fix any errors in the generated code by passing them back to the LLM. In the future, this product could be extended to generate code for the full build, train, test, and measure cycle so that researchers can ask for a set of models to be tested, tweak the generated code as needed, and rapidly evaluate the best model for their needs.

NeuroGuru - AI Virtual Learning Assistant

NeuroGuru is a Vertex - Generative AI-powered educational platform designed to help users around the globe learn about artificial intelligence (AI) in their native language. Leveraging the advanced capabilities of models like PaLM2, Codey, NeuroGuru provides a personalised and interactive learning experience that adapts to each individual's learning pace and style. The platform covers a wide range of AI topics including Machine Learning, Deep Learning, Natural Language Processing, and Reinforcement Learning. It's not just a passive learning platform - NeuroGuru is designed to be interactive, engaging users with quizzes, hands-on coding exercises, and advanced topic explorations based on their interests and goals. One of the standout features of NeuroGuru is its adaptive learning capability. Based on the learner's progress and responses, NeuroGuru adjusts the learning content, making recommendations for what to learn next. It also includes a community feature for learners to connect, discuss, and collaborate on AI topics. NeuroGuru also emphasizes convenience and accessibility, with voice assistant integration and cross-platform support ensuring a seamless learning experience across computers, tablets, and smartphones. Whether you are a beginner just stepping into the world of AI, or an experienced professional wanting to keep up-to-date with the latest AI advancements, NeuroGuru is your comprehensive, go-to platform for AI education.

Two minutes Online

The whole theme of our Project revolves around the concept of "Make Your Presence Online in 2 minutes in 2 dollars. It is generally cumbersome for a business person to seek persons or company for making their websites. It takes generally one week to make full fledge website. But with help of AI, we can accomplish this task in 2 minutes. User will have to visit the website and fill a simple form in which he will provide his business details. On submitting the button create website, the website code in zip folder will be created in 2 minutes that could be downloaded by client. Client will be able to see relevant web-pages and contents in regard to business on his/her website.

Vertex Code Analysis

We are presenting a tool for static code analysis, based on a large language model by Google. It can assist you with code review and improvement. AI can greatly enhance code quality and save developer's time. Remember, fixing and improving the code is probably the most time-consuming part of the job. Conventional static code analysis tools are mostly rule-based. They can assist in: * checking style; * checking types; * finding basic security issues. In contrast, an AI-based code analysis tool can: * find potential bugs, based on syntax and semantics; * suggest code optimisations; * promote best coding practices. We believe this project can also find its place in the field of products for developer efficiency. With rapidly improving large language models, organisations and individuals are willing to boost their productivity, and pay for subscriptions or pay-as-you-go services.

IntelliAudit

IntelliAudit is able to take a git repository containing smart contracts, either a full project with hardhat, truffle, brownie, etc or by selecting individual contracts from the repository and audit the code for vulnerabilities. IntelliAudit leverages slither in order to parse the smart contracts in a structured way and provides several functions to chatcode-bison in order to request data about a certain contract and its functions in order to be able to statically analyze the content and determine any vulnerabilities. IntelliAudit uses langchain in order to create an agent that keeps hunting continuously until it finds a vulnerability or has evaluated the full code base. A vector database with different smart contracts vulnerabilities that have been discovered in the past assists the agent in determining any vulnerable code.

Filters

AI Models

Data

Platforms