BridgeIt

• 

  Workflow of Bridgeit.

Inspiration

Our inspiration for developing BridgeIt originates from our collective experiences working in interdisciplinary fields such as marketing, technology, strategy, and research. Coming from computer science and data science backgrounds, we often struggled to understand and communicate complex concepts across different domains. For instance, when my research professor, a neurobiologist, explains intricate biological concepts, it requires extensive learning from research papers, resulting in significant latency in communication and understanding. More specifically, in enterprises such as FLOZ, our mentor Hanyang has experienced disconnects between his architectural teams and his engineering teams - resulting in roadblocks in project execution. As seen, these delays hinder progress, especially in enterprise environments where collaboration between diverse disciplines is essential. We envisioned BridgeIt as a solution to bridge these gaps, using AI to make information accessible, clear, and tailored to each user's expertise.

What it does

BridgeIt is a documentation app designed to facilitate seamless communication and knowledge sharing within large organizations. It uses AI to transform complex data into clear, role-specific insights, supports rich media, and enables real-time collaboration. The app integrates with popular tools like Slack, Microsoft Teams, Jira, and Confluence, making it a comprehensive solution for enhancing team productivity and efficiency.

How we built it

We built BridgeIt using Vertex AI on Google Cloud for robust and scalable AI capabilities. The frontend is a Swift app designed for a smooth and intuitive user experience, capable of understanding and conveying complex information effectively. We also utilized LangChain to enhance the app's natural language processing and understanding capabilities. The integration of these technologies ensures that BridgeIt delivers precise, context-aware insights tailored to each user.

Challenges we ran into

Building BridgeIt presented several challenges:

NLP and Fine-Tuning: We faced significant learning curves in natural language processing (NLP) and fine-tuning AI models. It required a lot of experimentation with different frameworks to achieve the desired accuracy and efficiency.

Hallucination in AI: Understanding and mitigating AI hallucination was crucial. We had to come up with strategies to minimize incorrect or misleading outputs generated by the AI models.

Integration and Security: Ensuring seamless integration with various third-party tools while maintaining data security and user privacy was a complex task.

User Interface: Creating a user-friendly interface that caters to users with different levels of technical expertise required extensive testing and iteration.

Accomplishments that we're proud of

We successfully attended numerous events at the hackathon while developing our product. Through these experiences, we gained valuable insights into our strengths and weaknesses. We had to adapt and optimize our approach to best support our team, which significantly enhanced our overall efficiency and effectiveness.

What we learned

We learned that there are significant gaps in the market regarding communication due to the prevalent differences in skillsets and technical understanding among team members as well as among the different teams that collaborate together. By engaging with startup like floz, lmnt etc in the startup valley, we realized the necessity of a tool like BridgeIt. Such a tool can greatly enhance workplace experiences by bridging communication gaps and making knowledge sharing more efficient and effective.

What's next for BridgeIt

Our goal is to make BridgeIt a resounding success by starting with a simple, yet effective, Chrome extension and Swift App. This will allow us to address immediate needs and gather user feedback early. As we evolve, we plan to develop a robust suite of tools that can be seamlessly integrated across various teams and departments. This suite will enhance communication, streamline knowledge sharing, and ultimately improve workplace experiences. Our vision includes expanding functionality, incorporating advanced AI features, and ensuring the tool adapts to the diverse needs of modern organizations.

Built With

• cloudrun
• datastore
• google-cloud
• langchain
• llamaindex
• swift