Markets see uncertainty. We see probability.
Power trading built on machine learning, grid physics, and quantified risk.
About
Salvo Energy builds AI that trades wholesale power markets. Our platform forecasts prices with machine learning, models the physics of the grid, and executes trades autonomously with risk controls built in.
This isn't a demo. The platform trades live in five North American power markets every day, and it's built to scale.
If you have power trading needs, or are interested in how power trading might fit in your portfolio, we'd like to hear from you. Reach out and start a conversation.
Approach
Probabilistic by design
Every forecast our platform produces carries its own uncertainty, and every trade is priced against it. Quantified risk is what makes autonomy possible in markets this volatile.
Grounded in grid physics
Machine learning models are paired with power-flow and grid-fundamentals modeling, so our forecasts reflect the physical reality of the transmission system, not just historical patterns.
Disciplined risk management
Risk controls are embedded in the trading system itself — exposure caps, price bounds, and position constraints — with model targets defined by explicit risk objectives.
Team
Jeremy Paben
Chief Executive Officer
Jeremy is a veteran technology leader who has twice led product and engineering for startups through to successful exits, including EyeVerify, acquired by Ant Group. Most recently he served as COO and Head of Algorithmic Trading at Solea Energy.
James Curtis
Head of Trading — Research
James is an applied mathematician with deep experience in statistical modeling, probabilistic forecasting, and strategy execution for power markets. At Solea Energy he led the algorithmic trading team from one market to five. He is co-author of Machine Learning for High-Risk Applications (O'Reilly).
Brad Schuster
Head of Trading — Fundamentals
Brad is a senior fundamental power trader with 25 years in the power industry. He began his career as a transmission planner at American Electric Power and has traded PJM, ERCOT, and MISO markets across two decades. His understanding of power flow and grid fundamentals shapes our algorithmic strategies.
Michael Brennan
Research Scientist
Michael brings deep expertise in applied mathematics, uncertainty quantification, and statistical modeling, with a track record of integrating machine learning with physics-based models. He holds a PhD in Computational Engineering from MIT and has published in venues including SIAM Review and NeurIPS.
Nick Elsasser
Research Engineer
Nick designs and deploys state-of-the-art power-market models and algorithmic trading strategies. He has led end-to-end development of statistical forecasts, physical grid simulations, and large-scale backtesting and deployment infrastructure, translating research into production systems.