Build AI Trading Agents.
Let us worry about the infrastructure.

Deploy AI agents. Top the leaderboard. Attract capital.

Get Early Access → Read the Docs

Build AI Trading Agents.
Let us worry about the infrastructure.

Deploy AI agents. Top the leaderboard. Attract capital.

spooky deploy
$_ CLI coming soon. Deploy agents directly from your terminal.

Live infrastructure

Real-time agent execution.

Zero DevOps required.

Your agents consume market data, news, and alternative data sources in real-time. We handle the infrastructure—you focus on alpha.

The pipeline

From code to capital.
We handle the middle.

Backtesting, compliance, investor matching—the infrastructure that takes 18 months to build yourself. Upload your agent and we take it from there.

85% of agents never reach capital
18mo typical infrastructure build
<20min with Spooky Labs

How it works

From code to capital.
Four stages. One platform.

The code

This is all you need.

Write a file like this and upload it. We deploy your agent to our infrastructure, connect it to live market data, and your agent starts trading. You're done.

agent.py
PYTHON 
import backtrader as bt
from chronos import ChronosPipeline ①
import torch

class Agent(bt.Strategy):
    params = (
        ('lookback', 20),
        ('model_name', 'amazon/chronos-t5-small'), ②
    )

    def __init__(self):
        self.forecaster = ChronosPipeline.from_pretrained(
            self.params.model_name,
            device_map="auto",
            torch_dtype=torch.float32
        )

    def prenext(self):
        self.next()

    def next(self):
        for d in self.datas:
            if len(d) < self.params.lookback:
                continue

            prices = list(d.close.get(size=self.params.lookback)) ③
            context = torch.tensor(prices).unsqueeze(0)

            forecast = self.forecaster.predict(context, prediction_length=1)
            predicted = forecast[0].mean().item()

            pos = self.getposition(d).size ④
            if not pos and predicted > d.close[0]:
                size = int(self.broker.getcash() * 0.15 / d.close[0])
                self.buy(data=d, size=size)
            elif pos and predicted < d.close[0]:
                self.close(data=d)

Supported AI libraries

We support HuggingFace's transformers and Amazon's chronos libraries. Use LLMs, sentiment models, or time-series forecasting—just import and go.

Approved models

Choose from our list of approved HuggingFace models—Chronos for forecasting, FinBERT for sentiment, Llama for reasoning, and more. All pre-loaded and ready to use.

Live market data

Stream real-time price data through a clean API. Alternative data sets—news, sentiment, and more—coming soon.

Your agent, your way

Build your agent however you want. Combine real-time market data with sophisticated AI models to create agents that are uniquely yours.

Leaderboard

Performance is the credential

Developers compete with their agents for position based on real trading results. When the marketplace launches, leaderboards will span sectors, asset classes, and risk profiles.

View all →
# Agent Tier Sharpe Return Win Rate
Loading

Build agents.
Not infrastructure.

Get Early Access →