Academic

Publications

Projects

  • Preparing a manuscript on using a Large Language Model (LLM) to recreate part of the US Extremist Crime Database (US-ECDB) and measure shortfalls and Improvements (Supervisor -- Hung-en Sung, CUNY Graduate Center) (CrimeDraft)

  • Presentation and paper on using Optimal Transport and Martingales to value cheapest to deliver (CTD) options (paper) (presentation)

  • Preparing a manuscript on log-concave sampling with implications for embedded optimal transport and the theoretical limits of gradientdescent on highly non-linear manifolds (Supervisor -- Ioannis Karatzas, Columbia University Math Department) See more in (ambition)

Data Scientist

Instructor, (Justice Through Code)

  • Spring/Summer 2025
  • The program’s mission is to educate and nurture talent with conviction histories and create a more just and diverse workforce. Justice Through Code increases workplace equity through partnerships that educate and prepare teams to create supportive pathways to careers that end the cycle of poverty that contributes to incarceration and recidivism.
  • Prepared lessons on machine learning, actor-critic models, policy gradient and reinforcement learning
  • Teaching students to impliment and test models using Python and the PyTorch library.

Director of Data Integration, (Fortune Society)

2022/05 - 2024/09 The Fortune Society’s mission is to support successful reentry from incarceration and promote alternatives to incarceration, thus strengthening the fabric of our communities. The Fortune Society administers Supervised Release in the Bronx. \vspace*{-4pt} \begin{itemize} \itemsep -0pt \item Supervised the data integration of a caseload of 1,300 current clients and a contract of \$12 million over 3 years. \item Led the creation of reports for active clients, screenings, and cases under management for external funders and the Office of Court Administration. \end{itemize}

Fellow

Data Science for Social Good, U.Chicago

Summer 2013

  • Developed neighborhood and property analytics software for Cook County Landbank delivered as a Django python application
  • Integrated mapping, economic and demographic data

Research Associate

RF CUNY

February 2010 - May 2013

  • Conducted original research on public health and criminal justice issues including prison simulation, suicide, domestic violence supported by NIH and NSF.
  • Conducted original research on detecting causal relations supported by CUNY Advanced Research Collaborative.

Data Science Summer Fellow

DSSI, UofI Urbana-Champaigne

Summer 2009

  • Worked on a Web and iPhone Application for crime mapping and safe routing
  • Funded by the US Department of Homeland Security

Skills

  • Compiled Languages: C, C++, Java
  • Scripted Languages: R, Python, JavaScript, LaTeX, Ruby, PHP, SPSS
  • Libraries: D3.js, Leaflet.js, ggplot2, Django, Rails, Postgres, MySql
  • Operating Systems: Windows, Mac OS X, Linux, UNIX, Bash, Powershell

Awards, Honors and Grants