Inspiration
Having AI models right on Chrome is a huge advantage in terms of reduction in inference latency and enables new client side applications. We tweet a lot on X.com about our projects, but often time don't get the opportunity to write down everything that comes to the mind on the main post due to free tier restrictions. Therefore we thought of creating a tool that summarizes the tweet within the free tier limits without compromising on the context.
What it does
1) Summarizes the user's tweet using 100 characters max to restrict the content within the free-tier limit. This helps user avoid the monthly subscription fee on X.
2) Finds relevant hash tags for the tweet if required by the user.
3) Translates the summarized tweet if required by the user.
How we built it
We used Chrome API and Google Gemini Nano models to create the chrome extension.
API's used - 1) Summarize API to summarize the tweet in under 100 characters. 2) Write API to find hash tags (worked better than Prompt API) 3) Translate API to translate the content to Spanish.
Challenges we ran into
1) Translate API often doesn't work. 2) Inserting the content directly into the page DOM also proved to be a pain due to the dynamic rendering of the X.com page.
Accomplishments that we're proud of
1) Successfully used more than 1 Built-in AI API to create the tool. 2) First time creating a Chrome Extension.
What we learned
1) Learned about the huge market potential of extension and workspace tools. 2) Learned about Chrome Extension development.
What's next for Save-On-X
1) Replace the Gemini Nano models with the Flash model for production. 2) Market fit validation with potential users.
Log in or sign up for Devpost to join the conversation.