Jason Flynn
Phoenixville,PA
Summary
Versatile Data Scientist and Quantitative Researcher with over 9 years of experience at the intersection of machine learning, predictive analytics, and financial research. Strong academic foundation in Applied Statistics, Economics, and Political Science, combined with a proven track record leading data science teams in developing production-level models and research pipelines. Deep expertise in statistical modeling, macroeconomics, commercial real estate (CRE) analytics, and AI, with a growing focus on generative AI and advanced ML techniques. Adept at bridging theory and practice to deliver actionable insights in high-stakes environments. Lifelong learner passionate about solving complex problems with data.
Overview
8
8
years of professional experience
Work History
SR. ASSOCIATE | QUANTITATIVE INVESTMENT RESEARCH
CBRE INVESTMENT MANAGEMENT
01.2022 - Current
- Lead data scientist on a team of quantitative investment researchers investing in public and private market Commercial Real Estate & Infrastructure at a Global market level.
- Work in close collaboration with portfolio managers to develop forward-looking investment tools that guide $110B in public & private market AUM
- Constrained portfolio optimization using quadratic utility functions, factor exposures, and factor-based covariance matrices.
- Portfolio allocation using Regime switching multi-factor models
- Portfolio allocation using econometric risk & return models and macro-economic scenarios
- Developed "Winning Cities" ranking methodology
- Econometric, statistical & machine learning models to enhance factor-tilting strategies.
- Used Generative AI to develop automated sentiment-checker using 10k SEC filings, earnings call transcripts and sell-side research. Employed GPT 4.0 API & Pinecone Vector Databases to achieve Question Answering
- Forecasts of cap rates, vacancies, and rent growth at the global sector level.
- Designed asset allocation framework tailored for institutional investor clients. This framework uses macro scenarios to forecast risk & return expectations. These forecasts serve as basis for constrained optimization tool allowing idiosyncratic portfolio allocation with risk targeting
- Mentored junior associates, fostering professional growth and helping them reach their full potential. Trained and supported new team members, maintaining culture of collaboration.
Assoc. Data Scientist & Data Engineer | Predictive Analytics & Macro Forecast Groups
Moody’s Analytics
01.2017 - 01.2022
- Lead developer of machine learning product to assess the risk of small & medium sized private enterprises
- This product won several awards, including a Fintech ‘Analytics Innovation Award’, and I was invited to speak as presenter at multiple professional conferences
- Model was constructed using a stacked ensemble framework incorporating several supervised machine learning algorithms
- Also involved in probability of default modeling related to Moody’s CreditEdge EDF product
- Responsible for automating & maintaining data pipelines for Moody’s Analytics Data Buffet
- This involved the construction of automated ETL pipelines to extract, QA, estimate and database third party economic data releases into Moody’s Data Buffet ecosystem
Education
MS - APPLIED STATISTICS
WEST CHESTER UNIVERSITY
West Chester, PA01.2021
BA - ECONOMICS
UNIVERSITY OF NORTH CAROLINA WILMINGTON
Wilmington, NC01.2016
BA - POLITICAL SCIENCE
UNIVERSITY OF NORTH CAROLINA WILMINGTON
Wilmington, NC01.2015
Skills
- Programming/Technical Skills: R, Python, SQL, AWS S3, Athena, Lambda, Sagemaker, Snowflake, Dataiku, Tensorflow, Keras, Huggingface, PineCone vector database, H2O, Fastest, caret, scikit-learn, tidyverse, pandas, Llama, OpernAI, Claude
- Machine Learning & Statistical Modeling: Random Forest, Gradient Boosting (XGboost, LightGBM, GBM, EBM), Support vector machines, various other regression & classification models, Generalized Additive Models, Time series forecasting (ANOVA, ARIMA, VAR/VEC, ARDL) Monte Carlo & Bayesian methods, Mean/Variance optimization, quadratic utility functions, etc
- Generative AI & Deep Learning: GPT3, GPT35-Turbo, GPT4, HuggingFace, GANs, Tensorflow, Keras, Pinecone vector databases, Langchain
- Financials: Portfolio optimization, Quadratic utility functions, Constrained optimization, robust estimation of covariance matrices, GARCH modeling, factor-tilt strategies
Timeline
SR. ASSOCIATE | QUANTITATIVE INVESTMENT RESEARCH
CBRE INVESTMENT MANAGEMENT
01.2022 - Current
Assoc. Data Scientist & Data Engineer | Predictive Analytics & Macro Forecast Groups
Moody’s Analytics
01.2017 - 01.2022
MS - APPLIED STATISTICS
WEST CHESTER UNIVERSITY
BA - ECONOMICS
UNIVERSITY OF NORTH CAROLINA WILMINGTON
BA - POLITICAL SCIENCE
UNIVERSITY OF NORTH CAROLINA WILMINGTON
Similar Profiles
Joshua KumarJoshua Kumar
Work experience at CBRE Investment ManagementWork experience at CBRE Investment Management
Olivera MarićOlivera Marić
Manager for retail leasing at CBRE SEE (Property management & retail leasing)Manager for retail leasing at CBRE SEE (Property management & retail leasing)
Grace RaminGrace Ramin
Finance Director, GWS Global & Asia Pacific at Global Project Management, CBRE Pte LtdFinance Director, GWS Global & Asia Pacific at Global Project Management, CBRE Pte Ltd