Experience

• 

  Intech

  • Massachusetts, United States

  • Data Science and Alpha Generation

    • May 2024 - Present
    • Massachusetts, United States

• 

  Acadian Asset Management

  • Boston, Massachusetts, United States

  • Portfolio Manager / Quant Researcher

    • Nov 2020 - May 2024
    • Boston, Massachusetts, United States

    Developed quant strategies and products in the Equity Alternatives Group, focusing on uncorrelated additions to the multistrat product. Employed machine learning, NLP, data science and a lot of coding. Also produced analytics, reports, and presentations for internal and external consumption.

• 

  Balyasny Asset Management L.P.

  • Boston, Massachusetts, United States

  • Senior Quantitative Analyst

    • May 2020 - Nov 2020
    • Boston, Massachusetts, United States

    Machine learning and NLP projects for equity portfolio strategies.

• 

  GMO

  • Greater Boston Area

  • Quantitative Analyst

    • Aug 2016 - May 2020
    • Greater Boston Area

    Quant equity investing research with GMO's Global Equity team. Primarily worked on developing a variety of stock selection alpha signals and enhancing existing ones.Used and actively investigated a plethora of techniques, including non-traditional, Data Science-y approaches. Special focus was on unstructured or, to put it more directly, really messy data, e.g. text, ownership, or investment flows.Technical lead/subject matter expert on several NLP, machine learning, and other projects on GMO's nascent cloud-based platform in the Python/Databricks/Spark/Azure ecosystem.

• 

  On Mini Sabbatical

  • Independent Researcher

    • Jul 2016 - Aug 2016

    Spent several weeks working on a backlog of personal scientific/coding/finance projects [and some fun stuff too] in anticipation of new position that commenced in late August, 2016.

• 

  Acadian Asset Management

  • United States
  • Financial Services
  • 400 - 500 Employee

  • Portfolio Manager, VP

    • Jan 2013 - Jul 2016

    I may have been Portfolio Manager on my business card but I spent most of my time doing research on quantitative equity forecasting models. I focused particularly on developing and improving alpha signals for bottom-up stock selection. Most of "my" factors were highly technical in nature. They spanned a range from classics such as momentum and reversal all the way through ones derived from unstructured text and other big-data-ish sources requiring NLP and other machine-learning treatment.My projects covered the full life cycle of factor development: theory and hypothesis formation, data acquisition and cleaning, factor code development and backtesting, written and verbal presentations to colleagues, and near-production-ready implementation. Post-production, I monitored factor performance and continued to investigate enhancements and (rarely, of course!) fixes. As tools I used mainly Python (including packages such as pandas, numpy, NLTK, scikit-learn) and SQL. I did dabble in some R and Matlab. Lately I was and remain quite interested in issues of visualization.

  • Investment Research Analyst

    • Aug 2011 - Jan 2013

  • Senior Software Engineer

    • Dec 2007 - Aug 2011

    My developer team was focused almost exclusively on supporting the core research and investment groups. As a result my projects cut across almost all of the areas of quant investment process from data gathering and analysis, through bottom-up stock selection signals, top-down models, combination of same into the overall forecast, and even some aspects of the risk model and portfolio construction.A "classic" task involved taking factor ideas from researchers and implementing them in production. In virtually all cases this required careful understanding and scrutiny of underlying theory and vetting of final results. It was an iterative process of identifying steps in the computational algorithm that required modification to account for data issues, atypical market segments or regimes, unexpected corner cases, etc. The ultimate production signal often looked significantly different from the original proposal.Our principal software tools here were SQL, Matlab, and C++.

• 

  Nuance Communications

  • Greater Boston Area

  • Senior Research Engineer

    • Mar 2004 - Nov 2007
    • Greater Boston Area

    At Nuance (formerly ScanSoft), as a member of the MREC group, I worked on the core Dragon NaturallySpeaking automatic speech recognition engine. Our team bridged "pure" research from speech scientists and application-focused tasks of product development groups. Crucially, our codebase was used by both. As such, it had to be all things: fast, ultra-reliable, flexible, and with a robust but intuitive API.The most significant part of my job involved vetting, modifying, and finally implementing proposed enhancements and additions to the statistically-driven models and algorithms that formed the recognizer. These included a number of real-time speaker adaptation techniques, expanded language models, improved decoding/search algorithms, etc.The engine was implemented in C++ and we also provided an API/user hooks in Python.

• 

  SpeechWorks

  • Greater Boston Area

  • Speech Synthesis Engineer

    • Apr 2001 - Jan 2003
    • Greater Boston Area

• 

  Openwave Systems

  • Greater Boston Area

  • Principal Software Engineer

    • Feb 1999 - Apr 2001
    • Greater Boston Area

• 

  Massachusetts Institute of Technology (MIT)

  • Research Laboratory of Electronics, Cambridge, MA

  • Research Associate

    • Jan 1995 - Feb 1999
    • Research Laboratory of Electronics, Cambridge, MA

    Conducted R&D on communication aids for the profoundly hearing impaired. Specifically, led the development of a prototype Automatic Speech Recognition-driven system producing supplementary visual cues to individuals using Cued Speech. In laboratory experiments this clearly improved the subjects' speech reception.

• 

  Karlsruher Institut für Technologie (KIT)

  • Karlsruhe Area, Germany

  • Postdoctoral Fellow

    • Oct 1993 - Dec 1994
    • Karlsruhe Area, Germany

    Research and development of an Automatic Speech Recognition system aided by automatic lip-reading using artificial neural networks.My own research focused on the image/video processing and feature extraction stages. However, I also co-ordinated the effort to integrate several projects into a single, multi-modal, human-computer interaction system. This included face location and tracking, speaker identification, image segmentation, automatic lip reading, and ASR. In addition I supervised several Bachelor's (Studienarbeit) and Master's (Diplomarbeit) theses.