Inspiration
The idea of Amazon and the relationship between different products as I shop online there for products influenced me to look into this idea and see how it could be improved.
What it does
The agentic app uses tools along with AQL, NetworkX, and cuGraph algorithms to solve problems, which in this case helps us understand more about the relationship between different products that Amazon offers to improve its services.
How we built it
The agentic app was built using Python in a Jupyter Notebook.
Challenges we ran into
Challenges I ran into were getting things set up initially as well as long wait times for certain programs to generate, such as for graphs. Some graphs did not generate with the correct information initially, so some trade-offs had to be made to ensure successful generation.
Accomplishments that we're proud of
Accomplishments that I am proud of is creating a functional app with an agent that answers queries, as well as being able to optimize the agent to allow more efficient and robust querying to solve problems.
What we learned
I learned more about working with databases and how AI can be used in beneficial ways to solve problems.
What's next for ArangoDB Agentic App
The app can be fine-tuned some more for deeper understanding of the queries we want to pass in to understand the relationship between Amazon products better.
Log in or sign up for Devpost to join the conversation.