2019-10-03 · 3 min read
This blog post is the first in a series that will spotlight some interesting contributions, features, or changes that members of the community have implemented for Hummingbot.
The first company we are featuring is Dolomite, a decentralized exchange built on top of the Loopring protocol, a purposed-built DEX protocol designed to achieve high-transaction throughput. The company has just launched their exchange, in which Hummingbot played the integral role of providing liquidity. The Dolomite team looked to Hummingbot as a more cost effective, DIY alternative to paying expensive market makers to provide liquidity for their exchange.
We are big fans of the Dolomite team and what they have accomplished, and they provide a great example of Hummingbot’s mission of open finance. Providing liquidity to newly established exchanges is a compelling use case for Hummingbot, helping to kick start new entrants and provide them with help to compete with existing exchanges.
The Dolomite team have been part of the Hummingbot community since the launch of our alpha program in March 2019. They have not only built a connector that allows Hummingbot to execute trades on their exchange, they have built a custom market making strategy.
The Dolomite team first built a connector for Hummingbot, allowing Hummingbot users to transact on Dolomite. The connector is able to connect to both Dolomite’s testnet as well as main net. Dolomite is already using the connector.
For Hummingbot, the connector is currently under final stages of code review, after which, it will be merged and made available to Hummingbot users (to track the progress, navigate here).
The Dolomite team implemented a variation of the pure market making strategy called price_target_market_making.
The main features of this custom strategy are:
External price source (CoinMarketCap) to set the strategy’s mid price: whereas the existing pure market making strategy uses the exchange order book to set mid price, using an external price source may be a preferred and more reliable alternative, particularly for less liquid, thin order book exchanges.
”Lazy” order evaluation: the strategy periodically evaluates each order and classifies each order as “OK”, “High”, or “Low”, depending on whether or not the order is still at a sensible market level. The strategy also cancels/adjusts individual orders that may have moved out of range for the market. Especially when running the strategy with a large number of multiple orders, this methodology more efficiently manages outstanding orders versus the default cancel all/replace all strategy of the base strategy.
Lastly, the Dolomite team also implemented a visualization into the strategy, showing order book depth, sizes at each price point. Below are some screenshots of the strategy in action, placing 25 orders on each side of the market, with increasing order sizes with increasing spreads:
Dolomite’s exchange is available at https://beta.dolomite.io. As an incentive for market makers, the exchange offers -0.1% market maker fees.
For developers interested in creating exchange connectors or custom strategies, please review the following resources: