DealMates

pitch

Inspiration

We were inspired by the challenges that many people face when trying to find the best deals or negotiate favorable terms—especially those with limited resources or experience. We wanted to create a tool that levels the playing field, using AI to automate negotiations for everything from second-hand goods to real estate and healthcare, making these opportunities accessible to everyone.

What it does

DealMates is a Chrome extension that uses AI agents to help users find the best deals on marketplaces and negotiate favorable terms for them. It automates the process of searching for items and services, comparing prices, and bargaining on behalf of the user. The technology can also be applied to various industries such as real estate, healthcare, and financial services, where negotiation plays a crucial role.

How we built it

We built DealMates using a combination of AI models fine-tuned for negotiation tasks, integrating them into a user-friendly Chrome extension. On the backend, we leveraged AI agent frameworks like Dendrite for automating negotiation strategies. Prem AI was used to finetune models for our negotiation and sentiment agent. We used Fetch AI to allow effective communication between the agents. We used Mistral 8x7b as our LLM model. The frontend was built using JavaScript and React to ensure smooth user interactions, and Groq's infrastructure was utilized for the quick inferencing capabilities.

Challenges we ran into

One of the biggest challenges was ensuring that the AI agents were able to negotiate effectively across different marketplaces, each with unique rules and pricing structures. We also faced difficulties in scaling the AI models for various applications, from e-commerce to real estate, while maintaining high performance. Another challenge was managing team resources and coordinating technical development within tight deadlines.

Accomplishments that we're proud of

We successfully developed a working MVP that allows AI agents to negotiate prices on behalf of users. Our AI agents are capable of handling complex negotiations, adjusting strategies based on real-time data, and delivering tangible savings for users. We're also proud of the versatility of our solution, with potential applications across industries beyond just e-commerce.

What we learned

We learned the importance of collaboration and the ability to pivot quickly when challenges arise. Developing AI models that can negotiate effectively across diverse sectors requires fine-tuning and constant iteration. Additionally, building a scalable backend infrastructure that can handle various use cases taught us how to balance performance and complexity.

What's next for DealMates

We plan to expand the platform’s capabilities by applying the technology to larger markets like real estate, healthcare, and financial services. We also aim to refine the AI agents further, improving their negotiation strategies and scaling them to handle even more complex transactions. Ultimately, we hope to create a powerful tool that democratizes negotiation for everyone.