-
Login Page
-
List of tabs available inside the application
-
Product Roadmap creator with just a feature prompt
-
Decision Logger with Vonage SMS API (Message is sent directly to the entered mobile number when decision is logged)
-
SMS received on my number when decision was logged
-
PRD generator with detailed information on problem statement, goals, requirements, metrics and timeline for the completion)
-
OKR fit analyzer tab gives a fit score with subject to OKR
-
Release Readiness check covers the gap and give recommendations
Inspiration
Every day, product managers are overloaded with decisions, documentation, and strategy alignment and all of which slow down momentum. I wanted to build an assistant that takes care of the “busy work” in product management so teams can focus on building. That’s how the idea for an AI Product Assistant was born.
What it does
The AI Product Assistant automatically logs product decisions, analyzes OKR fit, and generates PRDs on the fly. It also sends real-time updates via Vonage SMS so stakeholders stay in the loop instantly on decisions with no dashboards or emails required.
How we built it
The frontend is powered by Streamlit with a modular design for decision logging, OKR alignment, roadmap tracking, and PRD generation. I integrated Vonage's SMS API to deliver critical updates in real time. The backend uses Supabase for persistent memory, and the AI reasoning engine runs on an LLM accessed via the Incredible API.
Challenges we ran into
Vonage SMS API required precise configuration and error handling, and managing API secrets securely on Streamlit Cloud added friction. I also ran into issues with dependency conflicts and cloud deployment errors that required careful debugging and I'm still working on it.
Accomplishments that we're proud of
I was able to integrate LLM-driven reasoning, SMS notifications, and a structured product workflow in one app. The Vonage integration was a real highlight — it felt magical when that first SMS alert went through!
What we learned
I gained deep experience integrating AI workflows with external APIs like Vonage, learned about LLMs and incorporate AI reasoning and features, and securing credentials when pushing to git. Most importantly, I learned how to simplify complex product workflows through thoughtful automation.
What's next for AI Product Assistant with Vonage SMS
Next steps include adding SMS-based inputs (reply to log a decision!), integrating with Jira, and turning the assistant into a fully autonomous agentic system that learns from user behavior and optimizes decisions automatically.
Built With
- api
- authentication
- cloud
- llm
- python
- sms
- streamlit
- supabase
- vonage
Log in or sign up for Devpost to join the conversation.