%201.webp)
Aethon Fund has launched with USD $50 million in capital. The hedge fund was founded by George Kailas, who also started retail investor platform Prospero.ai.
The raise includes an anchor allocation in a separately managed account from a fund of funds, alongside commitments from other ultra-high-net-worth and institutional investors.
Kailas said the new fund draws on a research base built over two decades and on signal tools developed through Prospero.ai, which he founded in 2019. The platform has more than 20,000 monthly active users. Aethon said its signal library was informed by four years of live calls with more than 200,000 retail investors.
The fund will run several trading strategies in parallel, including long, short, mean-reversion, momentum and what it describes as stealth accumulation. Capital will shift toward the strategies performing best under current market conditions. Artificial intelligence systems will handle execution, while the investment team focuses on signal research and technology.
Each position is subject to preset rules for profit targets, stop-loss, trailing, and time stops, the firm said. No position will be held open indefinitely, and a variance risk premium overlay adjusts long and short exposure according to market regime classification.
Prospero link
Aethon's launch is closely tied to changes at Prospero.ai. Kailas is becoming Chairman of Prospero.ai and stepping back from day-to-day management, while Adam Plante, the company's longtime Chief Technology Officer, becomes Chief Executive Officer.
A percentage of Aethon's proceeds will be directed back to Prospero.ai. The arrangement is intended to support the retail platform's ongoing work.
"Most funds guess. We measure," said George Kailas, Founder and leader of Aethon Fund.
"Aethon is the product of twenty years identifying what the models don't know and building signals to close those gaps. We're not a quant fund with a bigger data budget. We're a fund where the signals themselves are the moat, refined through years of real-world feedback loops," Kailas said.
Kailas described the relationship between the two businesses as central to the fund's development. He said feedback from Prospero.ai users helped shape the underlying tools now used by Aethon.
"The Prospero.ai community doesn't just use our signals - they helped build them. Every call, every piece of feedback, every market they pushed us to cover made these tools what they are today. The best way I know to honor that is to leave the platform in the hands of someone who understands the mission as deeply as Adam does, and to give back to the community through the profits Aethon Fund will achieve," Kailas said.
Signal set
Aethon said its research stack includes nine daily signals across more than 2,000 stocks, a social sentiment system tracking conversations across Reddit, X and StockTwits, an analyst-rating model that weights price targets by recent forecasting accuracy, and a model that predicts return probabilities across stocks and exchange-traded funds over different time horizons.
The firm said these signals were tested in live market conditions rather than only in backtests. It described the surviving models as interpretable and predictive, and said the system uses agreement or disagreement between signals and strategies to adjust position sizes.
Kailas said Aethon also plans to disclose examples of how its signals work on specific stocks. That would mark an unusual level of openness in a sector where managers often keep research methods confidential.
"There is no industry precedent for a hedge fund doing this. That's the point. Most participants in our industry have treated their data as classified," said George Kailas, Founder and Leader of Aethon Fund.
"We're going the other way. We have tools that can help people understand what's actually happening in the market, and we're going to share them, where appropriate and meaningful," Kailas said.
Leadership team
Aethon has assembled a management team spanning trading, artificial intelligence, infrastructure, communications and risk management. Dave Lauer joins as Chief Technology Officer after earlier roles as a high-frequency trading quant and trader at Citadel and Allston. Ezi Ozoani becomes Head of AI after leading large-language-model and artificial-intelligence systems at Hugging Face.
Joe Bernstein has been appointed Head of Trading after eight years developing and executing systematic trading strategies at Tower Research. Erik Smolinski takes the role of Head of Risk Management, while Dave West becomes Head of Infrastructure. Dr Matt Carter will serve as Head of Communications.
Kailas's background includes starting work at a hedge fund at age 17 and selling a mortgage prediction model to a hedge fund managing USD $20 billion at age 25. The new fund is based in New York City.