Director - Insights & Analytics, Lenati, a marketing and sales strategy consulting firm. I lead a team of seven data scientists and market researchers to support marketing and sales strategies. We did data science projects at companies including Adobe, Foot Locker, Airbnb, and Lululemon. I also created new methodologies such as the Lenati Loyalty ROI Simulator, which used agent-based customer simulation to predict the value of a loyalty program design. (2016-2018)
Lead - Advanced Analytics, Promontory Growth and Innovation [Formerly Insights Results]. I lead a team that used data science to support the process improvement program. We found millions of dollars in savings by analyzing data, and created interactive tools to explore company costs. (2012-2016)
Strategy Analysis Specialist, The Boeing Company. I helped the Market Analysis team create the Current Market Outlook, an annual 20-year forecast of the airline industry. I used mathematical modeling and forecasting to supplement the team's industry expertise in predicting the future market size. (2010-2012)
Analyst, Vistaprint. I lead a team that created a tool that used statistical quality control and forecasting techniques to monitor sales data and assess if something was wrong with the website. I also redesigned the company internal forecasts to predict sales at the daily level. (2009-2010)
Strong: R (+Shiny), Python, SQL, Keras, Git, Excel, F#, C#, Power BI, Linux, AWS, Azure
The vehicle scheduling problem for fleets with alternative-fuel vehicles, Adler, J.D.,
Mirchandani, P.B., Transportation Science: March 2016.
This paper discusses the problem of scheduling a fleet of buses which use alternative-fuel vehicles and have a limited range they can travel before needing to refuel. Exact and heuristic methods are proposed, then they are tested on data from the Valley Metro bus network in the Phoenix Arizona metropolitan area.
Online routing and battery reservations for electric vehicles with swappable batteries,
Adler, J.D., Mirchandani, P.B., Transportation Research Part B: December 2014.
In this journal article, a system is devised to route electric vehicles as they randomly arrive in a system while trying to minimize the global travel times. An approximate dynamic programming approach is used, and the algorithm is tested on Arizona highway data.
Routing and Scheduling of Electric and Alternative-Fuel Vehicles, Adler, J.D., Arizona
State University Dissertation: April 2014.
My dissertation, which is a collection of several of the other publications listed here.
The Electric Vehicle Shortest-Walk Problem With Battery Exchanges, Adler, J.D., Mirchandani,
P.B., Xue, G., Minjun, X., Networks and Spatial Economics: January 2014.
This paper shows how the shortest-walk network problem is different when traversing a network with a range-limited electric vehicle. A solution is provided for the problem, and the problem is also solved with an additional constraint of a limited number of allowed stops.
New Logistical Issues in Using Electric Vehicle Fleets with Battery Exchange Infrastructure,
Mirchandani, P.B., Adler, J.D., Madsen, O., Procedia - Social and Behavioral Sciences: January
This paper gives an overview of some of the difficulties of solving classic operations research transportation networking problems with the additional constraint of using electric vehicles which can only travel a limited range before needing to recharge.
Over the past few years I have been doing a lecture tour, where I help people understand data science in industry. This includes helping students prepare for getting jobs, businesses strengthen their analytics offereings, and technical experts see new methodologies. My list of presentations includes:
Everyone keeps saying that there are tons of jobs for mathematicians in business and industry, but if you search on career boards for "mathematician" all you get are teaching jobs. There is a set of buzzwords that companies use to describe math jobs such as: analytics, operations research, and data science. In this presentation I discuss what steps students can take to make them great candidates for industry, how to find the jobs that have the interesting math problems behind them, and how to answer the age old question: should I go to grad school?
A student who has studied mathematics has many skills that can be useful in an industry setting, but often times seeing how a transcript can transfer into value to your company isn't straightforward. In this talk I will discuss how math students can help your businesses as well as how to find the students that will be a good fit for you.
Many factors go into the success of a data science team: finding the right project, hiring the right people, and using the right tools. In this presentation I go over some of the best practices around data science teams, examine case studies from successful projects, and talk about how to use the right tools to go from data on a server to a finished product.
Doing statistical analyses and machine learning in R requires many different components: data, code, models, outputs, and presentations. While one person can usually keep track of their own work, as you grow into a team of people it becomes more important to keep coordinated. This session discusses the data science work we do at Lenati, a marketing and strategy consulting firm, and why R is a great tool for us. It covers the best practices we found for working on R code together over many projects and people, and how we handle the occasional instances where we must use other languages.
My writing on an assorted set of topics.
Hiring data scientists (part 1): what to look for in a candidate
What are the technical and business skills required to be a data scientist.
Hiring data scientists (part 2): the perfect candidate doesn't exist
What are the most to least important skills, and what are the types of people who apply for data science jobs.
Hiring data scientists (part 3): interview questions
The questions I ask when interviewing candidates.
Hiring data scientists (part 4): the case study
The last part of my data science interview process.
Getting data science to work
How to think about doing predictive modeling in a business situation.
Choosing the right graduate degree for data science
How to compare the many options when looking for a masters to help you in data science.