AI case study

Public charging trust

by
Georgia Tech
Context

Georgia Tech partnered with Microsoft using Azure OpenAI Service to analyze global EV charging behavior data in 72 languages, automating the classification and predictive modeling of public charging experiences. They fine-tuned their AI with context prompts and reinforcement learning from human feedback, integrating it into research workflows to support policy development and infrastructure enhancement.

Results

Classification time reduced from 99 weeks to minutes; cost reduced by 3-4 orders of magnitude (approx. USD 0.006 per observation).

Results not reported in the source
Industry
Education
Published
August 14, 2024
Agent type
Data Agents
AI provider
Microsoft
Models/tools
Not disclosed
ICE score
567
The ICE framework in this database provides a quick way to assess the feasibility and potential impact of AI use cases, with higher scores signaling more actionable opportunities.

Impact: Potential benefits to the business.

Confidence: Likelihood of achieving expected results.

Ease: Simplicity of implementation in terms of resources and time.

ICE Score: Calculated by multiplying the component scores.

Note:
Each score is AI-generated based on available data and should be viewed merely as a general guideline for deeper exploration of the use cases.
Impact
9
Confidence
9
Ease
7

43

AI use cases in

Education

See All

Pensieve

Education
Use case
Personalized student support
Context

Pensieve integrated Anthropic's Claude models (Claude 3.5 Sonnet and Claude 3.7 Sonnet) to build AI teaching assistants that automate grading and provide 24/7 personalized tutoring. The system clusters student submissions to calibrate grading rubrics and transforms static PDFs into interactive worksheets while enforcing instructor-controlled policies. This integration streamlines course management workflows and enhances real‐time classroom analytics.

Models/tools

Super Teacher

Education
Use case
Building learning tools and making content
Context

Super Teacher integrated Anthropic's Claude to generate initial code for interactive educational tools and draft lesson plans for subjects ranging from pre-K to 5th grade. They embedded the AI into both software development and content creation workflows with mandatory human review to ensure safety and quality standards. This integration streamlined operational processes, enabling engineering and content teams to focus on higher-level development and creative enhancements.

Models/tools
...
1

Praxis AI

Education
Use case
24/7 academic support
Context

Praxis AI integrated Anthropic's Claude 3.5 Sonnet via Amazon Bedrock to create AI-powered digital twins of professors that serve as virtual teaching assistants. They implemented this by switching from GPT-4 to Claude using an LLM switcher and deploying the solution on a secure, scalable AWS infrastructure that supports multi-agent workflows and real-time conversational features. This approach enhanced round-the-clock personalized student engagement and alleviated faculty workload.

Models/tools
Explore industries

149

companies using

Data Agents

See All
Use case
Automated job title classification
Context

Aura Intelligence integrated Anthropic's Claude via Amazon Bedrock into its data pipeline to automatically classify over 200 million job titles and industry pairings from multi-language data, replacing manual lookups and fuzzy matching. They fine-tuned foundation models on proprietary datasets and leveraged AWS infrastructure, including SageMaker and prompt management, to automate QA, report generation, anomaly detection, and real-time hiring trend analysis.

Models/tools
...
2
Use case
Efficient engineering team management
Context

LaunchNotes leverages Claude in Amazon Bedrock in their product 'Graph' to transform engineering data into actionable insights. Graph functions as an ETL platform with Claude managing data pipelines, helping engineering managers understand development metrics, reduce incident identification time, automate updates, and generate customized release notes and technical documentation.

Models/tools
...
2
Use case
Faster data insights
Context

Snowflake integrated Anthropic's Claude into their platform to enable natural language queries on complex databases, allowing customers to extract insights without SQL expertise while maintaining security and governance.

Models/tools
...
2
Explore agents

226

solutions powered by

Microsoft

See All
Use case
Multilingual reading difficulties
Context

The school deployed Microsoft Reading Progress and Microsoft Immersive Reader, two AI-powered apps integrated within Microsoft Teams, to support individualized learning. The tools record reading sessions, provide real-time feedback on pronunciation and errors, and translate texts to enable comprehension across multiple languages. This setup streamlines classroom instruction, personalizes student learning, and reduces teacher workload.

Models/tools
...
2
Use case
Faster customer issue resolution
Context

Vodafone uses Azure OpenAI Service, Azure AI Foundry, Microsoft Copilot, and Azure AI Search to enhance their virtual assistant TOBi and develop SuperAgent to assist customer service agents. TOBi uses conversational AI to handle customer inquiries, while SuperAgent helps agents solve complex problems faster.

Models/tools
...
4
Use case
Faster oncology data extraction
Context

Ontada leveraged Microsoft’s Azure OpenAI Service Batch API and Azure AI Foundry to build its ON.Genuity platform, which processes 150 million unstructured oncology documents to automatically extract nearly 100 critical data elements across 39 cancer types. They integrated the new platform with their structured iKnowMed system using Azure Databricks for data ingestion and Azure Document Intelligence for text extraction, transforming manual chart review processes into an automated workflow that supports clinical decision-making and life science product development.

Models/tools
...
4
Explore AI providers

279

AI use cases in

North America

See All
Use case
Rigid support workflows
Context

Zendesk integrated OpenAI's models to create adaptive AI service agents that autonomously manage customer conversations and execute resolution tasks. They implemented a multi-agent architecture featuring task identification, conversational RAG, procedure compilation, and procedure execution agents integrated with existing support workflows through API calls and natural language procedure definitions, while providing real-time chain-of-thought visibility. This solution transitions from traditional intent-based bots to a hybrid model of scripted and generative reasoning, streamlining customer service processes.

Models/tools
...
2
Use case
Instant personalized support
Context

Chatbase integrated Anthropic's Claude into its multi-channel customer support platform to automate routine inquiries and deliver instant, personalized support across web chat, WhatsApp, Slack, and Instagram. They implemented the solution by embedding AI agents through APIs that integrate with business systems such as Stripe for order status checks and refund processing, while enabling customizable brand voice and multilingual communication with optional human oversight. This approach streamlined customer service workflows and provided advanced analytics to help teams quickly identify trends and optimize interactions.

Models/tools
...
1
Use case
Personalized student support
Context

Pensieve integrated Anthropic's Claude models (Claude 3.5 Sonnet and Claude 3.7 Sonnet) to build AI teaching assistants that automate grading and provide 24/7 personalized tutoring. The system clusters student submissions to calibrate grading rubrics and transforms static PDFs into interactive worksheets while enforcing instructor-controlled policies. This integration streamlines course management workflows and enhances real‐time classroom analytics.

Models/tools
...
2
Explore regions
Thoughts & ideas