Skip to main content

Ecosystem Vision

ClyptQ Trading Engine is a Trading Commerce platform that establishes trust between Buyers and Builders. Core Values:
  • Transparency: All strategy performance is verifiable and reproducible
  • Reliability: Backtest and live trading run on the same engine for consistent results
  • Scalability: Supports all strategies from AI-based to traditional algorithms
  • Accessibility: Buyers evaluate strategies without code, Builders access production infrastructure instantly

Trading Commerce Ecosystem

Builder (Strategy Creator)
    ↓ Develop & Test
    ↓ List on Marketplace

ClyptQ Trading Engine
    - Validates Performance
    - Ensures Reproducibility
    - Provides Infrastructure

    ↓ Browse & Evaluate
    ↓ Deploy with Confidence

Buyer (Capital Provider)

Builder Benefits

  • Production infrastructure (data, execution, monitoring)
  • Identical environment from backtest to live
  • Revenue generation via marketplace

Buyer Benefits

  • Access to verified strategies
  • Transparent performance metrics
  • Built-in risk management

Architecture Overview

ClyptQ consists of 4 layers:
  • Operator Protocol: Standard interface for all operations
  • TaggedArray: 4-field data structure (value, exists, valid, updated)
  • ComputeGraph: Dependency-based automatic execution
  • TradingDriver: System-wide orchestration
  • Signal: Alpha/Factor signal generation
  • Filter: Universe selection
  • Indicator: Technical indicators (40+ built-in)
  • Metric: Performance measurement
  • Semantic: LLM integration
  • TradingState: Portfolio state tracking
  • Execution: Order execution pipeline
  • Intention: Declarative order specification
  • Spec: Unified configuration system
  • Sources: Multi-source integration (crypto, stocks, macro)
  • Storage: Environment-based routing (dev/prod/kernel)
  • Validation: Data quality verification

Key Concepts

Operator Protocol

All operations follow the same interface:
class Operator:
    def compute(self, data, timestamp, context):
        """
        Args:
            data: Dict[input_id, TaggedArray]
            timestamp: Current timestamp
            context: Execution context

        Returns:
            TaggedArray with 4 fields
        """
Benefits:
  • Composable with consistent interface
  • Type safety guaranteed
  • Automatic dependency resolution

TaggedArray (4-Field System)

All data is represented with 4 fields:
TaggedArray:
    value: float      # Actual value
    exists: bool      # Data exists
    valid: bool       # Value validity (not NaN)
    updated: bool     # Updated at current timestamp
Benefits:
  • Explicit missing data handling
  • FFill distinction (valid=True, updated=False)
  • Consistent error handling

ComputeGraph

Strategies are represented as DAGs (Directed Acyclic Graphs) of Operators:
graph = ComputeGraph()

# Nodes
close = graph.add_node("close", FieldExtractor("close"))
sma = graph.add_node("sma", SMA(Input("close", lookback=20), span=20))
signal = graph.add_node("signal", MomentumAlpha(Input("close", lookback=20)))

# Automatic dependency resolution and execution
results = graph.compute(timestamp, data, context)
Benefits:
  • Automatic dependency resolution
  • Parallel execution optimization
  • Reproducible execution

Execution Modes

ClyptQ supports 3 execution modes:
Strategy validation with historical data
  • INSTANT mode: Perfect execution (no slippage)
  • LATENT mode: Realistic simulation (orderbook matching)

Why ClyptQ?

For Builders

  • Fast Development: Instant access to production infrastructure
  • Reliability: Backtest = Live (same code, same execution)
  • Monetization: Sell strategies via marketplace listing

For Buyers

  • Transparency: All performance metrics verifiable
  • Trust: Reproducible backtest results
  • Safety: Built-in risk management tools

Platform Benefits

  • Standardization: All strategies use the same protocol
  • Scalability: From single strategy to portfolio
  • Flexibility: Supports all strategies from algorithms to AI

Next Steps