SimpleFunctions provides a unified API that aggregates real‑time data from major prediction‑market platforms and delivers quantitative indicators such as implied yield, edge, and liquidity score. The service lets traders define natural‑language theses that are converted into actionable contracts and supports automated, risk‑gated trade execution via REST or CLI integration.
Funding
Funding not disclosed
Founders
Product
Problem
Traders and analysts must manually collect and interpret data from multiple prediction‑market platforms, which is time‑consuming, error‑prone, and lacks standardized indicators for automated decision‑making.
Solution
SimpleFunctions offers a unified API that aggregates real‑time data from major prediction‑market exchanges such as Kalshi and Polymarket, distilling thousands of contracts into concise, tokenized summaries. The platform provides quantitative indicators—including implied yield, cliff risk, expected edge, liquidity score, and time decay—to surface mispricings and market regimes. Users can define natural‑language theses that are automatically converted into causal trees of testable claims, which the system maps to relevant contracts and evaluates for edge detection. Built‑in risk‑gated triggers and a smart runtime enable autonomous execution of trades based on predefined conditions, while a CLI and REST endpoints allow seamless integration into existing trading stacks. Continuous 15‑minute monitoring, news scraping, and LLM‑enhanced search keep the system up‑to‑date without manual intervention.
Target Audience
Primary users are quantitative traders, hedge‑fund analysts, and developers building prediction‑market‑based strategies who need low‑latency data, actionable indicators, and automated execution capabilities.
Features
- One‑line API call that compresses ~18 000 active prediction‑market contracts into ~800 tokens, calibrated by real money outcomes
- Quantitative screen endpoint filtering markets by implied yield, cliff risk, expected edge, liquidity score, and time decay
- Regime classification that tags markets as maker‑friendly or taker‑friendly based on spread, depth, and flow patterns
- Thesis engine that transforms plain‑text hypotheses into causal trees, maps each node to contracts, and tracks edge size versus market price
- Edge detection with four gap types (consensus, attention, timing, risk premium) and execution‑ready sizing based on order‑book depth
- Declarative intent syntax with hard price triggers and soft LLM‑evaluated conditions (e.g., “only if oil > $95”)
- Autonomous runtime daemon that polls, evaluates, and fills intents, logs actions, and recovers from crashes
- Real‑time WebSocket and REST feed delivering sub‑second order‑book, trade, and candle data for integration with external systems