AI-Accelerated Sustainable Material Design for Bio-Inspired Sunscreens
Develop an intelligent agent that leverages AI and computational chemistry to discover and optimize novel, sustainable sunscreen pigment molecules. Inspired by the natural melanin pigments found in cephalopods, this challenge focuses on designing compounds with superior UV protection, enhanced biodegradability, and minimized environmental impact. Participants will build a multi-stage pipeline using Claude Opus 4.5 for de novo molecular generation, orchestrated by AgentFlow. Computational chemistry simulations, powered by Fireworks, will predict properties like UV absorption spectra, photochemical stability, and biodegradability. The agent will iteratively refine molecular structures based on these simulated outcomes, aiming for optimal performance and sustainability criteria.
What you are building
The core problem, expected build, and operating context for this challenge.
Leverage Claude Opus 4.5, AgentFlow, and Fireworks to design and evaluate novel, sustainable sunscreen pigments with optimized UV protection and environmental profiles.
How work is evaluated
The AI agent's performance will be evaluated on its ability to propose novel, sustainable sunscreen pigment molecules that meet specific UV absorption criteria, stability, and biodegradability. The evaluation will assess the pipeline's automation, the quality of generated structures, and the accuracy of property predictions.
Challenge datasets
Public datasets, plus the private datasets you have access to.
What you should walk away with
Develop multi-objective optimization strategies for chemical compounds.
Understand principles of bio-inspired material design and sustainable chemistry.
Apply Generative AI for de novo molecular structure generation and modification.
Integrate computational chemistry tools for accurate property prediction.
Utilize agentic workflows to orchestrate complex computational design pipelines.
Reference links and supporting material
Sample data for 3 tasks
Challenge setup
The AI agent's performance will be evaluated on its ability to propose novel, sustainable sunscreen pigment molecules that meet specific UV absorption criteria, stability, and biodegradability. The evaluation will ass...
Challenge input
JSON object with target properties (e.g., {"uv_absorption_peak_nm": 320, "biodegradability_score_target": 0.8})
Agent execution
No agent tool is listed for this challenge.
Evaluated output
JSON array of SMILES strings and initial predicted properties for generated molecules.
- Checks if the entire AgentFlow pipeline executes successfully without critical errors and produces valid JS...
- Verifies that Claude Opus 4.1 API was successfully called for molecular generation and refinement tasks.
- Confirms that Fireworks was correctly used to trigger and retrieve computational chemistry simulation results.
- UVAbsorptionEfficiency target: 0.9
- 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 the entire AgentFlow pipeline executes successfully without critical errors and produces valid JSON outputs at each stage.
- Verifies that Claude Opus 4.1 API was successfully called for molecular generation and refinement tasks.
- Confirms that Fireworks was correctly used to trigger and retrieve computational chemistry simulation results.
- Ensures that all generated and refined molecules are represented by chemically valid SMILES strings.
Metrics
- UVAbsorptionEfficiency · Target 0.9
- BiodegradabilityScore · Target 0.85
- StructuralNovelty · Target 0.75
Workflow tasks
- Task 1
Input: JSON object with target properties (e.g., {"uv_absorption_peak_nm": 320, "biodegradability_score_target": 0.8})
Output: JSON array of SMILES strings and initial predicted properties for generated molecules.
Execution harness: Python
- Task 2
Input: JSON array of SMILES strings from initial generation.
Output: JSON array of refined SMILES strings, each with comprehensive predicted properties (UV abs, stability, biodegradability) and a justification for refinement.
Execution harness: Python
- Task 3
Input: JSON array of refined SMILES strings and their properties.
Output: JSON object representing the single best optimized molecule, including its SMILES string, all predicted properties, and a rationale for its selection.
Execution harness: Python
Recipe freeze marker
No recipe freeze time is recorded.