Ohtani-san

Inspiration

We were inspired by the AT Protocol and platforms like Bluesky. We saw an opportunity to create a smarter way to track posts about specific athletes, moving beyond simple keyword filtering. The goal was to capture more relevant content for fans without the noise of irrelevant posts.

What it does

Our system feeds posts related to a specific athlete directly from the AT Protocol. It uses advanced filtering methods—more intelligent than basic keyword matching—to ensure that users see high-quality, relevant updates about their favorite player.

How we built it

We built the project on Google Cloud Platform (GCP), leveraging its robust infrastructure to handle real-time data processing. We integrated various APIs and developed smart filtering algorithms that analyze the content context, ensuring accurate delivery of posts about the athlete.

Challenges we ran into

We faced challenges in fine-tuning our filtering system to accurately distinguish relevant posts from similar, but unrelated content. Integrating with the AT Protocol and managing real-time data streams on GCP also required overcoming several technical hurdles.

Accomplishments that we're proud of

We are proud of creating a filtering system that outperforms traditional keyword-based approaches. Successfully integrating with the AT Protocol on GCP and delivering a more precise content feed for sports fans is a major milestone for us.

What we learned

Through this project, we learned a great deal about cloud computing on GCP, handling decentralized protocols, and developing intelligent filtering algorithms. The experience has deepened our understanding of both backend infrastructure and real-time data processing.

What's next for Ohtani-san

In the future, we plan to make the system lighter by adding support for multiple athletes and further improving the filtering model.

We were unable to improve the filtering model in time, but we were able to collect data, so we expect to be able to create a lighter model and reduce false positives.

Once the model becomes lighter, it will be possible to create a feed for individual MLB fans other than Shohei Ohtani, but this time we prioritized reducing running costs.

Our vision is to create a more dynamic and engaging feed that always keeps fans up to date with the most relevant and exciting posts.

Built With

• gcp
• gemini