Almanac is developed by a small, product-focused team led by Leo Lucian, with backgrounds in AI, data science, and prediction markets, operating independently within the Bittensor ecosystem.
What is Almanac?
Almanac is a decentralized prediction market platform built on Bittensor's Subnet 41. It aggregates sports predictions from independent AI models and data-driven participants into a collective intelligence layer, routing these signals through Polymarket to generate real trading outcomes. Rather than relying on a single centralized model, Almanac combines many independent forecasts to produce probabilistic predictions across major sports leagues and competitions. This approach demonstrates how incentive-aligned, decentralized AI within the Bittensor network can be applied to real-world prediction markets.
How Almanac works?
Almanac operates as a prediction market layer on top of Polymarket, powered by Bittensor's Subnet 41. Miners submit sports predictions by trading directly through the Almanac platform, while validators automatically track trading activity, verify outcomes using trusted external data sources, and compute performance scores over rolling evaluation periods. Predictions are assessed based on profitability and conviction-backed volume rather than simple accuracy alone. Incentives are distributed through Bittensor to reward consistently profitable performance, and the system continuously refines which signal providers receive the highest rewards through a two-phase optimization process.