The Tradegen roadmap is a summary of Tradegen's development and outline of Tradegen's future plans, which have been organized into four phases. Each phase is centered around a set of functionalities that will be delivered through continual updates.
While the phases of Tradegen will be delivered sequentially, the work for each phase happens in parallel, with research, prototyping, and development often in progress all at once.
The work of each phase is presented below. Here you will find an overview of the goals of the phase, as well as descriptions of the core functional components.
Tradegen is an asset management and algo trading platform, built from the ground up using a rigorous development model. The journey began in early 2021, with the vision of democratizing crypto trading. Several months and hundreds of hours of work later, Tradegen's first product - an asset management protocol - was shipped in November 2021, and phase 1 began.
The asset management protocol allows users to create, manage, and invest in pools (decentralized hedge funds) on the Celo blockchain. These pools can interact with whitelisted DeFi projects, allowing pool managers to utilize a pool's assets throughout the Celo ecosystem on behalf of the pool. This lets users gain passive exposure to active trading strategies, creates a new income stream for pool managers, and increases liquidity of other protocols in the ecosystem.
With the asset management protocol, Tradegen introduced two novel ideas: trading pool tokens as NFTs and a fully on-chain farming system. In addition to regular pools, Tradegen also lets users create 'capped pools' with a limited supply of tokens, each of which is an NFT that is tradable on the platform's marketplace. These NFTs can be deposited into farms to earn yield while staying invested in a pool. This farming system is unique because it is self-sustaining and fully on-chain. It's implemented as a 'hook' that gets called at the end of each pool transaction, updating a pool's weight in real-time based on unrealized profits and price performance over 2-week periods.
Following on from the asset management phase, phase 2 of Tradegen involves the development of custom infrastructure needed to power future algo trading products. Our custom infrastructure will include an application-specific Cosmos-based blockchain for running trading bots and a decentralized keeper network that relays bot performance data from our custom blockchain to leading blockchains supporting smart contracts, allowing developers to use the data in other projects (such as synthetic assets or prediction markets).
We decided to use an application-specific blockchain for running trading bots because it lets users verify the code and trades for each trading bot, gives us flexibility for scaling without having to compete with other applications, and allows us to remain decentralized. Our blockchain runs trading bots in simulated trades using price feeds from Band Protocol. Using simulated trades, instead of actual trades, leads to zero fees/slippage while preventing frontrunning, boosting the returns of existing strategies and allowing new strategies that were previously unprofitable after accounting for fees/slippage.
The main goal of this phase is to create a decentralized, verifiable, tamper-proof API for trading bot performance that updates in real-time based on live market data. Developers could use this API in synthetic asset protocols to tokenize trading bot performance, allowing users to invest in the strategy behind each trading bot.
After developing the infrastructure for running trading bots and relaying their data to leading blockchains, the next step is to build protocols that use the bot performance data. We plan to launch two protocols in phase 3: a protocol for creating trading bots, and a protocol for tokenizing the performance of trading bots.
The protocol for creating trading bots will initially launch on Cosmos, and would allow users to build trading bots, as well as the components that make up a trading bot (indicators and comparators for entry/exit rules). Each trading bot and indicator/comparator will be an NFT that is tradable on the platform's marketplace. A user holding a trading bot NFT can collect fees whenever a developer uses that bot's performance data in an application. Whenever a user creates a trading bot, and raises the minimum funds (through a crowdloan) needed to register the bot, it will be submitted for approval through a decentralized voting system. Trading bots with the most votes each week will be published to the platform, allowing them to run on our custom blockchain.
We will also launch a synthetic asset protocol (on Celo) that tokenizes trading bot performance data from our blockchain, letting users invest in trading bots by buying tokens. To build this protocol, we will use our keeper network to relay bot performance data to an oracle contract on Celo every few minutes to create a price feed. This protocol will be similar to existing synthetic asset protocols, but instead of using a price feed for stocks/cryptos, we will use a price feed for trading bot performance.
By the end of phase 3, Tradegen will have created the foundation for fully decentralized algo trading. Users will own, and be able to monetize, most components of the platform.
Phase 4 will see the expansion of Tradegen's protocols onto other leading blockchains supporting smart contracts, as well as a wider variety of possible trading bots. The trading bots in phase 3 have limited capabilities because they can only trade one asset, use their max position size on each trade, and only work with price data in entry/exit rules. In phase 4, we plan to expand the capabilities of trading bots by allowing them to specify an asset with each simulated trade, use fractional position sizes or trade on leverage, and support other data in indicators/comparators other than price (such as an asset's volume or metrics for the broader crypto market).