Gender-Equity-Navigator

By leveraging Retrieval-Augmented Generation (RAG) technology,Gender Equity Navigator draws from a rich repository of global reports to provide tailored insights on gender equity challenges.

Comment

Inspiration

This project was inspired by the need for better tools to understand and promote gender equity. I aimed to create a solution that could transform data and reports into clear, actionable insights for organizations and individuals working to close gender gaps.

What it does

Users can ask complex questions about gender equity topics, such as diversity policies, gender pay gaps, and leadership representation, and receive clear, customized recommendations and information. By combining AI-driven analysis with Retrieval-Augmented Generation (RAG) technology, the tool empowers users to make informed, data-driven decisions that support gender equality initiatives.

How I built it

I used Google Cloud Storage buckets to store PDF reports, Vertex AI Vector Search for index storage , gemini as llm with the guidance of LlamaIndex documentation, and Mesop to build an app that lets users ask questions and receive recommendations on gender equity. Google Cloud Run served as the platform for deploying the Docker container repository. By integrating Retrieval-Augmented Generation (RAG) with data from reports and articles, the tool can provide both quick answers and more comprehensive reports, offering insights tailored to user queries.

Challenges I ran into

Creating Indices for Data: Setting up and managing a storage-efficient indexing system in Google Cloud was challenging and time-consuming due to storage limitations.

Authentication: Managing Google Cloud's authentication system was initially difficult, as it required learning to handle permissions and security configurations effectively.

Learning Google Cloud: As a first-time Google Cloud user, exploring its tools and capabilities involved a steep learning curve.

Accomplishments that I'm proud of

Accomplishments that I'm proud of include successfully building an AI-driven tool that can deliver insightful recommendations on gender equity, integrating diverse data sources into a cohesive platform, and overcoming technical challenges along the way. I'm especially proud of navigating the Google Cloud environment for the first time, setting up a robust indexing system.

What I learned

I gained valuable experience in using AI to process information and provide targeted recommendations. The project deepened my understanding of the various aspects of gender equity across different regions and industries, showing me the power of data-driven action on these issues.

What's next for Gender-Equity-Navigator

I plan to integrate an SQL database to handle numerical data and extract tables directly from reports, enabling even more precise insights, especially around statistics and key metrics. Another priority is adding the ability to generate full reports as outputs, making it easier for users to get comprehensive summaries. I also envision using Google Pub/Sub for real-time data extraction, allowing the tool to provide updates as new information becomes available. Implementing these features globally will take this project to the next level, making it accessible and impactful worldwide.

Built With

Share this project:

Updates