# Roadmap Overview
Yala’s development follows a staged roadmap, with each phase expanding intelligence, validation, and system scope. This staged approach allows Yala to maintain continuous public value output and narrative relevance with minimal initial system complexity.
### Early Stage: Establishing the First Fair-Value Agent
In the early stage:
* Yala conducts closed, internal testing of its first fair-value AI agent
* A steady stream of probability estimates is released publicly via Yala’s official X account
* The focus is on methodology, calibration, consistency, and probabilistic reasoning
These early outputs demonstrate how Yala approaches fair-value estimation and lay the foundation for more advanced capabilities.
### Mid Stage: Public Launch of the Fair-Value AI Agent
In the mid stage, Yala publicly launches its first fair-value AI agent.
**Agent Scope**
* Designed for price-prediction markets and risk-neutral valuation
* Operates as a verifiable, measurable system
* Performance is continuously tested in real markets
**Primary Signal Sources**
* Historical trading data
* News-driven event analysis
* Smart-money tagging information
* Social-media-based sentiment dynamics
**User Inputs**
Users interact with the agent via a structured input format:
* Market type (sports events or crypto markets)
* Target condition (price, direction, or range)
* Time horizon (future timestamp)
**Outputs**
* A probability estimate representing fair value for the specified condition
**Live Validation**
* The agent operates in a controlled live environment
* Manages approximately $1,000–$10,000 in capital
* Executes trades autonomously. In validation environments, probability signals are translated into position-sizing decisions using predefined deviation thresholds between fair value and market-implied probabilities.
* Validates fair-value logic under real-world conditions with strict risk limits
### Late Stage: Multi-Agent Fair-Value System
In the late stage, Yala evolves into a comprehensive multi-agent fair-value system capable of generating explainable probability assessments across markets, signals, and event types.
The system expands into a coordinated swarm architecture, where a Supervisor (Orchestrator) Agent coordinates specialized Worker Agents to produce multi-factor fair-value outputs that integrate both risk-neutral and subjective probabilities.
**Expanded Scope and Outputs**
Users can submit queries across any asset or event category, including crypto, equities, elections, esports, and macro outcomes, along with a future time horizon.
The system generates:
* Probability density functions (PDFs)
* Multi-factor fair-value curves
* Confidence intervals
* Distribution shapes
These outputs provide a holistic probabilistic view beyond single-point estimates.
**Key Capability Expansions**
Compared with earlier stages, the late-stage system introduces:
* Subjective fair value alongside risk-neutral pricing, incorporating sentiment, macro, and contextual signals
* Multi-market coverage across the full prediction-market landscape
* Multi-factor intelligence, integrating options data, market sentiment, ETF flows, and macro events
* Multi-agent swarm architecture, enabling parallel analysis and coordinated signal aggregation
**Supervisor and Worker Agents**
The Supervisor (Orchestrator) Agent manages query parsing, task assignment, execution monitoring, and final aggregation.
Worker Agents include fair-value modeling, data collection, sentiment analysis, smart-money analysis, event tracking, options analysis, simulation, and decision aggregation agents.
**Advanced Capabilities**
The late stage also introduces private-information adjustment, encrypted data handling, strategy-gated prediction vaults, autonomous capital allocation, and tokenized agent monetization.
By this stage, Yala functions as a fair-value operating system supporting agent-driven forecasting, coordination, and prediction-market infrastructure at scale.
\
\
\
---
# Agent Instructions: Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://docs.yala.org/prediction-market-intelligence/roadmap-overview.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.