● Covered by 1 source · 1 reportLow impact1 neutral
Engineer creates Stonkfly, a simulated fruit fly brain that day trades cryptocurrency
Key points
- Stonkfly simulates a fruit fly brain with 166,700 neurons for crypto day trading.
- The virtual fly receives dopamine for profitable trades and aversion for losses.
- It trades without leverage or short selling, using real BTC-to-USDC data.
- The project is open-source and available for macOS and Linux with specific requirements.
Simulated Brain for Crypto Trading
An engineer from Coinbase developed Stonkfly, an open-source project that simulates a male fruit fly brain to perform cryptocurrency day trading. The simulation involves a brain with 166,700 neurons and 25.6 million connections, which interprets candlestick charts to make buy, sell, or hold decisions for various cryptocurrencies.
Reinforcement Learning Mechanism
The simulated fly's trading decisions are guided by a reinforcement learning mechanism. A rising portfolio value triggers a 'dopamine rush' in 15 cells as a positive reinforcement signal, while losses activate two aversive cells. Trading fees are also counted as losses. The simulation does not incorporate pain or emotional mechanisms, and the fly cannot use leveraged positions or short selling.
Visual Input and Trading Parameters
The virtual fly perceives the candlestick graph as a 320x180 display across its left and right eyes, with photoreceptor cells receiving RGB pixel values. The fly 'thinks' and acts every 500 milliseconds, with market data refreshing every 60 seconds. It can place orders up to $10, with a maximum of 24 orders per day.
Project Scope and Availability
The author states that the project primarily illustrates the link between input mechanisms, visual signals, and synaptic changes, rather than proving actual trading ability. The project is available for download on GitHub for macOS and Linux, requiring 16 GB of RAM, Python 3.11, and a C++ 17 compiler. It defaults to paper trades with a virtual balance of $100, using real BTC-to-USDC data.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
Primary sources
GitHub nftechie/stonkflyReporting from
A Coinbase engineer developed Stonkfly, an open-source project that simulates a fruit fly brain with 166,700 neurons to day trade cryptocurrency based on candlestick charts. The simulation uses positive reinforcement (dopamine rush) for profitable trades and negative reinforcement for losses, without employing leverage or short selling. This project demonstrates the connection between visual input, neural changes, and trading decisions, though the author cautions against interpreting it as actual trading ability.