Quantitative hedge fund applying machine learning to financial prediction and portfolio optimization
Voleon is a machine-learning-driven hedge fund founded by two ML scientists, now operating a research-heavy organization where doctorates dominate the technical staff. The tech stack—Python, Pandas, Airflow, Spark, Iceberg, Trino, Flink, Dagster, Kubernetes—reflects a mature data engineering operation built to handle large-scale financial datasets and live trading systems. Current hiring velocity is accelerating across research (13 open roles) and data (11), while active projects center on algorithmic execution quality, market prediction, and feature engineering, indicating ongoing investment in both model sophistication and production reliability.
Voleon is a quantitative hedge fund based in Berkeley, CA that applies statistical machine learning to investment management. The firm was founded in 2007 by machine learning researchers and is led by a CEO with a Ph.D. in Computer Science from Stanford and a Chief Investment Officer who is a statistics faculty member at UC Berkeley. The organization combines an academic research culture with emphasis on scalable systems architecture. The team is structured around research and data disciplines, with secondary support from engineering, finance, and operations. Voleon operates trading strategies and portfolio management systems across multiple asset classes.
Core stack: Python, Pandas, Apache Airflow, Apache Spark, Iceberg, Trino, Apache Flink, Dagster, Kubernetes, Bazel, Java, Go, C/C++. Infrastructure and collaboration tools include Smartsheet, Jira, Confluence, Workday, Slack, Zendesk, ServiceNow, Jamf Pro, Intune, Polars.
Active projects span financial market prediction, portfolio optimization, live trading productization, feature engineering for investment models, data pipeline production, algorithmic execution quality improvements, and trading strategy model development across asset classes.
The Voleon Group's technology stack, projects, and hiring signals are inferred from public hiring and company data — career pages, public listings, and company web presence — then clustered and de-duplicated. Figures are estimates that refresh over time. Read our full methodology →
This is not an official vendor or customer list. It is a technology-adoption signal inferred from public data, intended for B2B research.