Introduction

I have three principles in my approach to the development of new courses and curricula for students in the built environment. My first principle is to have a symbiotic relationship between teaching and research, where those domains of my academic work compound to build better outcomes for students and contribute to the fields of data science, real estate, and planning. Second, I believe in listening to students and stakeholders in person and through data. I’m an active user of data to inform decisions on my curriculum, but also on the ‘what’, ‘how’, and the ethics of ‘what’ I teach. Third, I prioritize the human experience in my courses: whether the class is stimulating; does it facilitate excellent technical outcomes in data science, finance, planning, or design; how might students apply learnings in the practice of city making. These three principles combined have led to numerous courses and classes, but also to my pedagogy of how to integrate the teaching of complex concepts from finance and statistics to data science and development. Below, I expand further on my principles with details of courses I have designed and implemented, from their original pique of inquiry, to how they work, and what they accomplished.

Circular Feedback Between Research and Teaching

The first principle in my pedagogy is that research drives my teaching and my teaching drives my research; they are symbiotic and necessary for them both to exist. My research and that of my team’s at the MIT Real Estate Innovation Lab has inspired numerous classes, seminars and workshops for graduate students and professionals in the built environment. Multiple courses in my portfolio started with a simple research question: for example, in DesignX, we asked “how do we develop an entrepreneurial venture accelerator program for students in the built environment?” This question led to papers, courses and workshops. As a second example, in our endeavour to value the strategic design and technology interventions of real estate, we asked “what are technologies for the built environment and how do they become a part of cities?” This question led to a graduate course, professional workshops, a technology for the built environment data science webtool – The MIT Tech Tracker, and now research papers. Often, the deployment of technical infrastructure assisting one aspect of research blossoms and restimulates others. For data science and cities, we focused on minimizing bias in my financial forecasting models to help guide strategic climate change and autonomous building interventions. To do this, I originated, financed and developed a geometric, geospatial, and relational database composed of hundreds of urban and real estate datasets. This led to renewed questions of how we would use big and wide data in new applications of machine learning for urban data science, and the birth of no fewer than three new courses and multiple published research papers. Research questions are necessary for understanding the theory and practice of planning and real estate. They help guide my pedagogy. However, they also advance the nascent use of urban analytics and data science in its practice and applications for a better built environment.

In this way, I have grown to teach multiple courses firmly based in theory and urban data science. My training is that of a financial economist. I teach courses in core finance, real estate finance, corporate finance, entrepreneurial finance, project development finance, and fixed income securities. However, I have also taken the time to invent my own courses in data science and machine learning, technological change and innovation, as well as be a core faculty member of studios in urban design and real estate development. Since my journey into teaching began the freshman year of my PhD, I have learned the power of teaching to students across multiple backgrounds, methods, and disciplines so as to help push forward the application and practice of real estate finance and development.

Listening to Students and Stakeholders

Perhaps one of the most important lessons that I have learned in practicing data science is that to be a good data scientist, one must be a great listener. There is evidence within the real estate development and planning domains that a leader’s project experience sets the benchmark for performance and policy. Unfortunately, listening to one expert’s opinion may lead to bias. For example, what if that professional has only dealt in a few events or experiences related to the policy, design, or development strategy their stakeholders are interested in implementing. No one person can command the scale of information on events that a large database can. This brings me to the second principle in my pedagogy, which is to listen attentively and carefully to our stakeholders — and that means working to collect data that helps us to answer relevant questions. How do we minimize bias by listening to more stakeholders, events, or outcomes in the problems that we are trying to resolve? My approach to urban analytics holds that there are signals in data that help us decipher what the urban environment and the human experience need.

At the same time, I’m also not naive; data science is a tool for interpretation, and not an oracle. Our field of urban analytics is at the beginning of the data science journey and we have yet to collect, analyze, and assess the full diversity of data that compose the multiplicity of experiences in a city. We only need to look at evidence of racial or gender bias in the built environment to help us see that our picture is far from complete. My work on data design, cultivation, and aggregation has taught me much about our unconscious mindsets in how we approach questions. I firmly believe that our first act must be to listen, absorb, and reflect with an arsenal of diverse thinking, expertise, and analysis.

Course Design

Education is an experience. At premier institutions that are working to shape the built environment of the 21st century, I have seen first-hand that the pedagogical silos between the domains of planning, architecture, urban design, or real estate development are not useful. Our students are transdisciplinary agents that are challenged with upcoming crises of climate change, inequality, automation, and the next pandemic. Thus, my third principle in teaching is one of scaffolding comprehension and skills over time. The subjects I teach — data science and finance — are subjects most students have little-to-no practical experience in graduate architecture and planning colleges. Both topics can seem highly esoteric, and so I have spent the last decade deconstructing and rebuilding their core concepts in ways more accessible to design and planning students. I make great use of the pedagogical practice of knowledge scaffolding; building upon simple concepts with layered strategies of content and practice over time. First, I employ multiple formats of engagement, like discussion forums, finance studios, data science charrettes, and diagramming workshops to provide a diversity of encounters and perspectives to the problems students are solving. Problem sets may test targeted application of comprehension, but my challenge is to create the designers and storytellers who invent the next stages of the built environment. Second, while delivering content at multiple levels, I leave students with a core takeaway — a so-called knowledge kit. Teaching in the built environment fields means teaching both theoretical foundations as well as real-world applications, so that students can continually reapply knowledge gained onto their foundational concepts. My students leave with coding templates, story-telling kits, data gathering strategies and technology bundles that are open to redeployment in their future practice. Third, I ultimately don’t know where my students will land, so I write and build content with my students that helps facilitate their future goals for various audiences and situations. My students have gone on to lead companies, execute projects within large architecture, consulting and real estate development firms and some have gone on to get their PhDs. My role as an educator is to deliver content and facilitate opportunities that enable them to go forward and develop more fully their desired impact on the built environment.

Outcomes and Forecasts

My training to become a teacher was developed over years. My first teaching practice occurred at an institution that applied problem based learning, which is normally used in medical school settings to deconstruct and diagnose problems as well as prescribe solutions. I have since blended this with the studio framework of a design and planning program, and the problem set structure of an engineering school. This has led to numerous measurable and non-measurable outcomes. My development of the MIT Real Estate Innovation Lab led to the teaching and supervising of over 56 students over five years. I have supervised over 65 master of science thesis. I have co-published with my students for professional and journal publications with over 30 white- and journal papers. I have designed seven graduate-level courses in my career and I have taught 15, over a range of topics including urban analytics, technology and quantitative finance. Below is an appendix of courses that I have designed or co-designed in the past three years.

I am currently designing two new courses. The first focuses on real estate finance for designers and planners. I want to use the techniques necessary for teaching design to reconstruct the traditional pedagogy of real estate finance. My goal is to have hundreds of students who can freely employ tools from planning, design, and finance to cultivate the built environment that they want. The second is an urban planning and development seminar that identifies deal structures for social inclusivity, with the intention of training cohorts of professionals, faculty, and communities of color in real estate development finance and analytics. My teaching goal for the next decade is to expand my knowledge in data science and finance to as many students in graduate and professional courses, to shift the inclusivity of these disciplines and help combat the built environment’s greatest challenges.

Appendix A:

Graduate Course Selection

Data Science and Machine Learning for Real Estate

At MIT I have designed multiple courses for graduate students that have been taken by planning, architecture, media, and real estate development students. Starting as a summer lab workshop in 2016, my current full semester course “Data Science and Machine Learning for Real Estate” was the first of its kind to help graduate students from planning, architecture, and real estate development approach the foundational concepts of data science, practical uses of econometrics in urban analytics, and the applications of machine learning in big and wide data sets.

My training in financial econometrics led to me quickly understanding that many problems in urban policy, real estate development, and design cannot be resolved using statistical practices applied in big data contexts. In some instances, we apply big data and its tools to urban problems. However, the variation in stakeholders, outcomes, and problems to be resolved requires approaches that can address both big and wide data problems. Wide data requires the application of analytical techniques to numerous data sources, identifying cooperative linkages between datasets, and ultimately the ability to connect outcomes and features to these multiple sources in various ways. The MIT Real Estate Innovation Lab built a geometric, geographic, and relational database of New York City for that purpose. Students practice analytics and machine learning techniques on wide data in urban and real estate. This course has been met with tremendous success, reaching students across the architecture program at MIT and Harvard.

Technological Change and Innovation in the Built Environment

There has been substantial hype about the nascent development of the proptech industry. With good reason, this jolt of innovation has led to massive investment in the digitization of urban, civic, architectural design, and real estate development practices. As important as these industries are to track and understand, my work seeks to frame changes in innovation using the academic field of technological change. This strategic focus on technologies for the built environment — as opposed to firms that apply technology — helps planners, architects and developers think more in line with the time horizons of their urban conditions.

To do so, I created a course called “Technological Change and Innovation in the Built Environment” that works with students to scout, track, analyze, and forecast the development and use of technologies being developed in the built environment. This course is built off of the methodology developed to produce knowledge graphs of technologies applicable to city making and building design, and provides engaging interactive experiences across invention hubs around the world to learn about the technologies that are just being invented, those that are ready to be picked up by the burgeoning proptech sector, and the arrival of technologies as staples in our everyday lives. The end result is an educational web application called the MIT Tech Tracker, built in conjunction between the MIT Real Estate Innovation Lab andJLL, and tested and developed throughout this course.

Real Estate Finance Bootcamp

Many students with a planning and design background do not understand the language and environment of finance and capital markets. For the MSRED program at the Center for Real Estate, I have worked with SA+P students extensively on a finance bootcamp that provides foundational concepts in finance, and translates these concepts into words and techniques that they are familiar with in planning and design. This is an intensive three week bootcamp taught daily, that helps students move from zero finance background to solid foundations for extensive coursework.

DesignX – Venture Design Curriculum

Graduate students in planning, design, and development are seeking ways to make an impact on the challenges the built environment faces. From climate change and resilience, income and housing inequality, social injustices in policy and practice, and market and policy failures of real estate development to name just a few. As a co-founder and Head of Research for DesignX, I worked with Professor Dennis Frenchman, Director Gilad Rosenzweig, and Faculty Director Professor Svafa Gronfeldt to develop the first and iterative approaches to teaching venture design to students. Principal to developing this course was research to understand how venture accelerators work, what they do, and how they lead to the success of businesses. We designed our accelerator and incentive program around this research. Currently, there are over 50 firms that have originated from the program, and a total of over $50 mln raised in Seed, Series A, and Acquisition funding.

Appendix B:

Professional Education Selection

GetSmarter – Data Science for Real Estate

My first online course for students around the globe is with the digital education company GetSmarter. This course is the first ‘Data Science for Real Estate’ of its kind. The objective of the course is to take professionals in the built environment or in data science and give them the foundations to understand what data is for the built environment, how you ask a question that can be answered using data, and applications of urban econometrics and machine learning to answer questions for stakeholders. Students walk away with code, data science tool kits, access to an urban analytics colloquium, and a network of professionals who are joining the application of data science in real estate.

MIT Professional Education – Data Science for Real Estate Development

Teaching professionals of varied age and professional backgrounds is critical for the advancement and diffusion of urban analytic techniques. If professionals within the industry are unable to even converse with their data scientist or understand their objectives, then management and professional adoption of urban analytics will be further slowed. This course teaches real estate professionals the practice of data science and its applications in real estate development using core methodology developed in the MIT Real Estate Innovation Lab.