About Unravel

Our Team & Ambitious Mission

Mission

Unravel is a quantitative investment research firm providing high-performance, institutional-grade, market-neutral factor portfolios based on sound statistical and behavioral edges.

We ingest vast, diverse set of exogenous datasets, creating academically inspired & proprietary factors — designing bespoke, multi-factor portfolios with excellent risk-adjusted returns (Sharpe Ratio of 2-3+). Exploiting cross-sectional alpha, with zero beta or market exposure.

Our public and proprietary research provides unprecedented transparency to make:

  • The most volatile and risky assets investable
  • Portfolios more resilient by exposing their vulnerability to exogenous risk factors
  • Quantitative strategies easier to implement and deploy with model portfolios and expert guidance

Unravel is built by a team of experienced quantitative analysts, software engineers and data scientists. We're pragmatists, committed to delivering results that are testable, repeatable, and transparent.

Leadership & Team

  • Mark Aron Szulyovszky

    Co-Founder

    Mark combines probabilistic thinking, technological excellence with proven execution, scaling his last tech company to $12m ARR. Post-exit, he's managed systematic trading strategies through his family office for the last 5 years. He co-founded Unravel to institutionalize & scale his quantitative approach.

  • Daniel Szemerey

    Co-Founder

    Daniel brings deep statistical expertise from his MSc Architectural Computation (UCL) and a track record of building data-driven ventures—from founding a supply-chain startup to helping 150+ Medical AI companies through his GE Healthcare powered accelerator. He co-founded Unravel combining technical depth with entrepreneurial execution.

  • David Parazs

    Quantitative Researcher

    Quantitative researcher with a chemistry background and systematic trading expertise. David earned a Research Chemistry degree from Tokyo Institute of Technology and worked as a Senior Data Scientist at WorldQuant Predictive. He specializes in statistical modeling and systematic strategy development.

SMA Partners

Unsigned Research Ltd. is our trusted Seperately Managed Accounts partner. Based on half a decade of digital asset management they bring deep factor research infrastructure with battle-tested execution and monitoring.

  • Charles Fried

    Unsigned Research

    Charles is a serial software entrepreneur with a proven track record in deep-tech ventures (Additive Flow). Frustrated by poor risk-management practices in crypto, he co-founded UR and devoted to developing sophisticated, systematic hedge fund strategies with proper risk controls and institutional-grade execution.

  • Dr. Alex Nestor-Bergmann

    Unsigned Research

    Alex holds a PhD in Applied Mathematics and was a Research Fellow at the University of Cambridge. With 10+ years in quantitative research and machine learning, he leads strategy design and risk systems with scientific rigour.

Advisors

We're building alongside experienced practitioners who share our vision in transparency, reproducible research and opportunities in multi-factor crypto portfolios.

Their involvement reflects a shared belief in our methodology and the significant potential we're unlocking together. We value their strategic perspective as we continue expanding our factor catalog and institutional partnerships.

  • Bo Zhang

    Advisor

    Industry leader in driving institutional growth, investment strategy evaluation, and treasury management. Bo has worked on treasury and investments with a variety of projects and family offices (e.g., Odos, Merit Circle, Function, and more), and has driven allocation of over $2B in capital within the blockchain space. Previously, Bo was a multi-asset portfolio manager in JPM’s Chief Investment Office, helping manage $600B+ in client assets.

  • Kris Longmore

    Advisor

    Kris transitioned from water engineering to trading, where he has worked for over ten years. He co-founded Robot Wealth in 2015, a systematic trading education platform serving 400+ members across six continents. Currently operates his own family office, deploying systematic strategies across multiple asset classes including equities, futures and cryptocurrency markets.

Origin Story

In 2020, Mark exited his previous company, and started his family office with the help of Daniel and a small team of quants. Together they deployed a wide range of quantitative (cross-sectional, long/short) strategies exploiting predictive relationships from exogenous data. Originally focusing on systematic macro, the two realized that there are more opportunities in digital assets and shifted all attention to market-neutral, cross-sectional crypto portfolios.

Unravel is proud on it's reproducible infrastructure and transparency: displaying a handful of cross-sectional factors on the website and auditable, runnable code that helps replicate Unravel's results and show transparent factor analysis.

Unravel is a manifestation of our passion for research and our belief that a well-defined discovery process can uncover many uncorrelated sources of alpha.

Office

114A Friedrichstraße
Berlin
10117
Germany

22 Berners St
London
W1T 3LP
United Kingdom

Contact

Timeline

Our Research & Development path so far.

Released Unravel

We publicly launched the research platform uncovering the exogenous predictive factors that matter most for any asset.

API released for partners

Launched a private programmable interface enabling clients to build systematic strategies using our core technology.

Packaged infrastructure for re-usable internal and external workflows

Standardized our technology stack to support both in-house research and partner funds.

Ranking and discovery of linear relationships

Built tools to automatically identify and prioritize significant linear patterns in financial data for more effective analysis.

Interpretability layer with human-in-the-loop step

Created systems that combine algorithmic insights with human expertise, ensuring models remain transparent and understandable.

Initial research into linear & non-linear forecasting methods with exogenous data

Began exploring advanced forecasting techniques that would enable us to surface exogenous predictive factors reliably.