Lecture Notes

Date: [Insert Date]

Introduction

• While waiting for others to join, casual discussion on weather in different regions.
• Attendees from various locations including India, East Africa, etc.

Lecture Start

• Focus on AI Project Management.
• Topics Covered: Project management in AI projects, real examples, group assignment details.

Key Points

Project Management in AI Projects

1. Goals of AI Projects:

   • Reduction of costs
   • Growth in revenue
   • Increase in customer satisfaction

2. ROI in AI Projects:

   • Importance of understanding the return on investment
   • Examples of ROI benefits like process efficiency and faster service deliveries

3. Types of Objectives:

   • Diverse project categories: BFSI, Healthcare, Supply Chain

4. Challenges and Considerations:

   • Identifying business problems suitable for AI
   • Training leadership and teams in AI and analytics
   • Securing budget and infrastructure
   • Piloting small scale projects before full implementation
   • Continuous monitoring and retraining of models

Group Assignment

1. Group Task: Create a business case using AI methodology.
2. Presentation Criteria:
   • Explain business case and its current solution
   • Describe how AI can improve the case
   • Discuss ROI and impact of implementation
   • Consider corner cases to be cautious of
3. Duration:
   • 15 minutes presentation, and 5-10 minutes Q&A
4. Logistics:
   • Groups to decide internal presenters
   • Any domain/topic choice allowed

Technical Discussion on AI Data Infrastructure

1. Data Sources Integration:
   • Internal and external data sources (e.g., credit ratings, KYC, etc.)
   • Methods: Batch and real-time data processing
2. Business Rules and Data Quality:
   • Importance of data preprocessing and transformation
3. Storage and Processing Flow:
   • Ingest raw data, apply business rules, process and transform data
4. Analytical Data Mark and Application:
   • Finalized data stored for analytical purposes
   • Potential use of AI models and automation for ongoing tasks
5. Query Layer:
   • Use of dashboards and natural language querying for data insights

Use Cases and Examples

1. BFSI Applications:
   • Predict loan delinquency
   • Asset restructuring evaluation
2. HR Analytics:
   • Predicting employee turnover and retention
3. Fraud Detection:
   • Using models to detect fraudulent transactions in real-time
4. Synthetic Data: Creation and ethical considerations in using AI-generated data for training models.

Closing and Reminders

• Ensure submission of the first assignment, if not done.
• Q&A session clarifying any doubts and offering additional insights.

Next Session<br>

• More examples of AI use cases in various fields.
• Continued discussion on practical AI solutions.