Dynamic Ad Pricing & Inventory Optimization
This challenge tasks you with building an advanced agentic system for dynamic mobile ad pricing and inventory optimization. Leveraging Claude Opus 4.1, your agents will analyze real-time market trends, predict ad performance, and intelligently adjust pricing and inventory allocation to maximize publisher revenue and fill rates. The core of this system will be a MCP-enabled framework to seamlessly integrate with simulated ad exchanges and external data APIs. This project emphasizes creating a robust, autonomous workflow where specialized agents collaborate to achieve complex business objectives. You'll design agents capable of continuous learning and adaptation, utilizing hybrid reasoning to respond to market fluctuations and identify optimal strategies. The goal is to demonstrate a cutting-edge application of generative AI in a fast-paced commercial environment, pushing the boundaries of automated decision-making.
What you are building
The core problem, expected build, and operating context for this challenge.
Build a CrewAI-based multi-agent system with Claude Opus 4.1 and MCP for dynamic mobile ad pricing and inventory optimization.
How work is evaluated
The evaluation will assess the completeness of the agent system, the correctness of MCP integration, and the effectiveness of the dynamic pricing and inventory optimization in a simulated environment.
Challenge datasets
Public datasets, plus the private datasets you have access to.
What you should walk away with
Design and implement a multi-agent system using CrewAI for ad inventory management.
Evaluate the system's performance against simulated market conditions and revenue metrics.
Integrate external data and simulated ad exchanges using the Model Context Protocol (Model Context Protocol).
Develop dynamic pricing and allocation strategies powered by Claude Opus 4.1.
Reference links and supporting material
Sample data for 3 tasks
Challenge setup
The evaluation will assess the completeness of the agent system, the correctness of MCP integration, and the effectiveness of the dynamic pricing and inventory optimization in a simulated environment.
Challenge input
Python code file (`agents.py`)
Agent execution
No agent tool is listed for this challenge.
Evaluated output
N/A
- Checks if CrewAI agents and tasks are correctly defined and can be initialized.
- Verifies that agents can successfully call tools integrated via MCP.
- Ensures the simulation script runs without errors and produces a report.
- Simulated Revenue Increase target: 0.2
- Python execution harness
Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.
View technical recipe
Configured tools
No tool records are attached.
Evaluation contract
- Checks if CrewAI agents and tasks are correctly defined and can be initialized.
- Verifies that agents can successfully call tools integrated via MCP.
- Ensures the simulation script runs without errors and produces a report.
Metrics
- Simulated Revenue Increase · Target 0.2
- Average Inventory Fill Rate · Target 0.9
Workflow tasks
- Task 1
Input: Python code file (`agents.py`)
Output: N/A
Execution harness: Python
- Task 2
Input: Python code file (`mcp_tools.py`)
Output: N/A
Execution harness: Python
- Task 3
Input: Python script (`run_simulation.py`) with configuration for market conditions.
Output: JSON report containing simulated revenue, fill rate, and agent decisions over time.
Execution harness: Python
Recipe freeze marker
No recipe freeze time is recorded.