Global AI Chip Supply Chain Resilience
Address the critical challenge of maintaining a resilient global AI chip supply chain. This challenge involves building a multi-agent system that simulates key nodes in a supply chain, monitors geopolitical and market events, and dynamically proposes strategies to mitigate disruptions. The system will utilize Gemini 2.5 Pro, particularly its 'Deep Think' mode, for advanced analytical reasoning and scenario planning to predict potential bottlenecks and identify alternative sourcing. AutoGen will be employed to orchestrate a decentralized network of 'Supply Chain Node' agents (e.g., Manufacturer Agent, Logistics Agent, Procurement Agent), each responsible for a part of the simulated supply chain. These agents will use hybrid reasoning (instant assessment of news, deep analysis of impact) and integrate with simulated ERP and logistics APIs to track inventory, production, and shipments, adapting their plans in response to real-time RAG-fed geopolitical and economic data.
What you are building
The core problem, expected build, and operating context for this challenge.
Address the critical challenge of maintaining a resilient global AI chip supply chain. This challenge involves building a multi-agent system that simulates key nodes in a supply chain, monitors geopolitical and market events, and dynamically proposes strategies to mitigate disruptions. The system will utilize Gemini 2.5 Pro, particularly its 'Deep Think' mode, for advanced analytical reasoning and scenario planning to predict potential bottlenecks and identify alternative sourcing. AutoGen will be employed to orchestrate a decentralized network of 'Supply Chain Node' agents (e.g., Manufacturer Agent, Logistics Agent, Procurement Agent), each responsible for a part of the simulated supply chain. These agents will use hybrid reasoning (instant assessment of news, deep analysis of impact) and integrate with simulated ERP and logistics APIs to track inventory, production, and shipments, adapting their plans in response to real-time RAG-fed geopolitical and economic data.
Shared data for this challenge
Review public datasets and any private uploads tied to your build.
What you should walk away with
Master AutoGen for setting up and orchestrating a decentralized multi-agent conversation framework, defining agent roles (e.g., Manufacturer, Logistics, Procurement, Risk Analyst) and their communication patterns.
Utilize Gemini 2.5 Pro, specifically its 'Deep Think' mode, for advanced causal reasoning, complex problem-solving in supply chain optimization, and generating detailed impact assessments of disruptions.
Implement a RAG system to continuously update agents with real-time news feeds (simulated), economic indicators, geopolitical events, and trade policy changes relevant to the global chip supply chain.
Develop a hybrid reasoning pipeline: agents use 'instant' reasoning (Gemini 2.5 Pro's faster modes or similar) for continuous monitoring and trigger 'Deep Think' for in-depth analysis of high-impact events.
Build MCP-style tool integrations for agents to interact with simulated enterprise systems like ERP (Enterprise Resource Planning) for inventory and production data, and TMS (Transportation Management System) for logistics tracking.
Design graph-based risk assessment models (conceptual or using a library like NetworkX) that agents can leverage to visualize and analyze interdependencies and vulnerabilities in the supply chain.
Implement adaptive planning mechanisms where agents can dynamically re-route shipments, adjust production schedules, or propose alternative suppliers in response to unforeseen events.
Explore techniques for agent self-correction and continuous learning within the simulation, enabling agents to refine their decision-making processes over multiple simulated cycles.
How this agent runs
The evaluation will assess the system's ability to identify, analyze, and mitigate disruptions in a simulated AI chip supply chain, focusing on the quality of proposed solutions and the efficiency of adaptation.
Challenge input
JSON object with 'initial_supply_chain_state': object, 'disruption_event': { 'type': string, 'location': string, 'impact': string, 'details': string }
Agent execution
The configured agent processes the input under the challenge policy.
Evaluated output
JSON object with 'impact_assessment': string, 'mitigation_plan': { 'actions': string[], 'affected_nodes': string[], 'estimated_recovery_time': stri...
- Verify active participation and communication among defined AutoGen agents during simulation.
- Ensure simulated ERP/logistics tools are correctly invoked and their outputs processed by agents.
- Mitigation Strategy Effectiveness target: 4
- Python execution harness
View technical recipe
Configured tools
No tool records are attached.
Evaluation contract
- The evaluation module defines the checks.
Recipe state
This is a preview. The configuration can change before the evaluation recipe is locked.
Run this agent on your dataset
Versalist can run this agent on your behalf with your data. Tell us about your dataset and the result you need.
Discuss your dataset[ok] Wrote CHALLENGE.md
[ok] Wrote .versalist.json
[ok] Wrote eval/examples.json
Requires VERSALIST_API_KEY. Works with any MCP-aware editor.
DocsFind another challenge
Jump to a random challenge when you want a fresh benchmark or a different problem space.