Generative AI Applications Hackathon winner built AI assistant for energy conservation meetings

News highlights 0

In alignment with the global ESG initiative, businesses are increasingly paying attention to energy management.

However, without specific expertise, most people have difficulty keeping abreast of energy efficiency issues. Fortunately, generative AI can help. Advantech provided the challenge for the smart energy management division of the AI Wave: Taiwan Generative AI Applications Hackathon, held on May 18-19, 2024 by DIGITIMES, challenging the participating teams to utilize generative AI to revolutionize user experience of smart energy management.

Advantech also provided resources and data for the teams to use. The team named "Playing a Part in AI", comprising De-Ru Yu and Si-Yu Tsai from 8 Interactive and Ting-Ni Wu from CTBC Securities, created AI Agent, a virtual expert guiding energy conservation meetings.

The team aimed to help managers new to energy conservation tackle challenges as they lack the experience to ask the right questions, sort through the data, and perform analyses. The team stood out from a group of strong competitors and won the honorable mention award, voted unanimously by all the judges.

Generative AI has demonstrated its strong capabilities in a multitude of application scenarios. Businesses are now striving to harness the power of generative AI.

To enable one-key data analysis and efficient energy management, Playing a Part in AI developed a solution that uses generative AI to look up data and prepare a presentation before a meeting, provide strategy suggestions and insights after the meeting, and regularly track records and personalized reports. It can help managers unfamiliar with energy conservation quickly grasp key points in a meeting.

Playing a Part in AI built its energy conservation meeting AI Agent using the Anthropic Claude 3 Sonnet model in Amazon Web Services (AWS) Amazon Bedrock, a fully managed service that makes high-performing foundation models on Amazon Web Services (AWS), coupled with AWS Lambda, a serverless compute service, and Amazon DynamoDB, a fast NoSQL key-value database service.

Before a meeting, managers only need to input the meeting topic, and the AI Agent will analyze a great amount of data and generate charts to help managers identify the key points. After the meeting, managers can use their mobile phones to upload data to the platform for the generative AI engine to sort, analyze, and archive. For regular tracking purposes, the AI Agent can send notifications with customized reports and charts according to user settings to help managers stay on top of project status.

The Playing a Part in AI team members come from various backgrounds encompassing engineering, marketing, and business planning. The team attributed their success to the comprehensive training they received at the AWS generative AI workshops. It brought them on the same page and allowed them to communicate without barriers so that they could complete the project and win the award.

The team named

The team named "Playing a Part in AI" won honorable mention for their AI Agent, a virtual expert for energy conservation meetings