Inspiration
We wanted to solve one of the most frustrating parts of product management manually assigning tickets. PMs spend hours balancing workloads and skills, so we built an AI assistant to handle it intelligently.
What it does
PM Ticket Manager uses GPT powered reasoning to automatically assign development tickets to the right developers based on workload, experience, skills, job title, and availability. PMs can review, approve, or reassign tickets with full transparency and visual insights.
How we built it
We built the backend with FastAPI and SQLite using OpenAI’s GPT 4o mini for intelligent assignments. The frontend was developed with React, Vite, and Tailwind CSS for a clean, responsive UI. Data is stored and visualized in real time using charts powered by Recharts and Axios for API communication.
Challenges we ran into
Tuning the AI prompts for consistent JSON responses, balancing workloads fairly across different developer types, and integrating parallel processing for speed were major technical hurdles. Managing synchronization between the backend database and the React frontend also required careful coordination.
Accomplishments that we're proud of
We created a fully functional AI driven ticket system with transparent reasoning, workload visualization, and real time updates. The tool can simulate realistic team environments using CSV data, making it perfect for hackathons or real PM workflows.
What we learned
We learned how to integrate GPT models into real applications, manage multi parameter AI logic, and balance explainability with automation. We also gained experience building full stack architectures that merge AI reasoning with human control.
What's next for PM_Ticket_Manager
We plan to integrate Jira and GitHub APIs, support multiple teams, enable user authentication, and introduce advanced analytics and reporting. Long term, we aim to train the AI on historical ticket data for predictive workload planning.
Built With
- api
- python

Log in or sign up for Devpost to join the conversation.