Home > Tasks > Reasoning > GeneticOptimizationTask

GeneticOptimizationTask

Iteratively evolve and perfect text through genetic algorithms. Uses mutation, crossover, and fitness-based selection to refine prompts, documentation, or messaging.

Category: Reasoning Iterative Evolution Multi-Generation
GeneticOptimizationConfig.json JSON 
{
  "initial_text": [
    "The product is good for developers."
  ],
  "optimization_goal": "Professional and persuasive",
  "num_generations": 5,
  "population_size": 6,
  "mutation_strategies": [
    "rephrase", 
    "simplify",
    "elaborate",
    "emphasize"
  ]
}
→
Evolution Analysis UI Markdown Render

Generation 5: Best Variant

Score: 94.5/100 (+22.1 improvement)

"Cognotik empowers engineering teams to automate complex reasoning workflows with industrial-grade precision and iterative refinement."
Clarity 92/100
Impact 96/100

Live Results Showcase

Explore actual artifacts generated by this task, including evolution logs and final markdown reports.

Execution Configuration

Field Type Default Description
initial_text * List<String> - The initial text(s) to optimize (seeds for the algorithm).
optimization_goal * String - The criteria for success (e.g., 'technical accuracy').
evaluation_weights Map<String, Double> Clarity: 0.35, Conciseness: 0.25... Weights for different scoring criteria.
constraints List<String> [] Additional context or constraints for optimization.
num_generations Int 5 Number of evolution cycles to perform.
population_size Int 6 Number of variants maintained per generation.
selection_size Int 2 Number of top candidates kept for the next generation.
mutation_strategies List<String> rephrase, simplify, elaborate Strategies used to generate variations.
enable_crossover Boolean true Combine traits from multiple top-performing candidates.