Build a CrewAI & OpenHands Epidemiological Outbreak Response Team
Public health agencies require automated multi-agent simulation tools to quickly model pathogen transmission dynamics and policy interventions. Build a CrewAI orchestrator that uses OpenHands code execution agents to run compartmental SIR models and generate outbreak mitigation strategies.
What you are building
The core problem, expected build, and operating context for this challenge.
Orchestrate CrewAI agents collaborating with OpenHands code execution to model public health epidemiological outbreaks.
How work is evaluated
Evaluates CrewAI crew performance in generating epidemiological simulations and policy outputs.
Shared data for this challenge
Review public datasets and any private uploads tied to your build.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
effective_r_reduction
Ensures effective R value under intervention is strictly lower than base R0.
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
simulation_accuracy
Accuracy of differential equation solving relative to standard ODE solutions. • target: 0.98 • range: 0-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Design specialized agent roles (Epidemiologist, Data Scientist, Policy Advisor) in CrewAI
Leverage OpenHands agentic sandbox execution to write and run SciPy ODE epidemiological models
Compute R0 transmission metrics and simulated hospitalization curves under varying policy conditions
Synthesize clear public health operational directives supported by model predictions
Reference links and supporting material
Parameter set containing transmission constants, host population demographics, and intervention efficacy estimates for 15 outbreak scenarios.
How this agent runs
Evaluates CrewAI crew performance in generating epidemiological simulations and policy outputs.
Challenge input
JSON with r0_value (float), population (int), and policy_intervention (string).
CrewAI
Coordinates specialized role-playing sub-agents.
OpenHands
Executes programmatic modeling code in isolated sandboxes.
Evaluated output
JSON containing estimated_r_eff (float), peak_hospitalizations (int), and policy_advice (string).
- Ensures effective R value under intervention is strictly lower than base R0.
- Accuracy of differential equation solving relative to standard ODE solutions. • target: 0.98 • range: 0-1
- Benchmark: EpiModel Policy Benchmark
- Simulation Accuracy target: 0.98
- 1 public reference case
- Python execution harness
- Python sandbox (unavailable on Versalist)
View technical recipe
Configured tools
- OpenHands · Required
- OpenHands · Required
- CrewAI · Optional
- Qwen2.5-72B · Optional
- OpenHands · Required
Evaluation contract
- effective_r_reduction · Weight 1
- simulation_accuracy · Weight 1
Recipe state
This is a preview. The configuration can change before the evaluation recipe is locked.