Inspiration
Developers spend way too much time crying in front of their Kanban Boards because their vocab is limited to Python libraries. We want to take that pain away for them... so that they can spend more time crying over their code instead.
What it does
TicketMaster auto-generates and de-duplicates tickets based on simple task prompts.
How I built it
We used Streamlit for the frontend and AWS + Qdrant and OpenAI APIs for our backend.
Challenges I ran into
Pydantic is awful. Just awful.
Accomplishments that I'm proud of
All our backend Lambdas and API Gateway routes deployed and worked on first try. It took one team member's 8 yrs of nginx experience to do this.
What I learned
Stop using backend for frontend (Streamlit/Flask).
What's next for TicketMaster
We want to do a dependency graph that will show you ticket dependencies (order of optimal completion) at a glance. Possibly using a GraphDB just because we can
Built With
- amazon-web-services
- docker
- javascript
- openai
- python
- qdrant
- streamlit
Log in or sign up for Devpost to join the conversation.