Michael Harper work email
- Valid
Michael Harper personal email
My dream is to combine my statistical modeling, analytics, and data science experience with my passion for sports.Technical skills15 years of experience as a quantitative researcher deriving actionable insights from complex data. Built cloud-based data processing pipelines in python, R, and SQL. Trained machine learning and statistical models using Keras library. Created dashboards in Tableau. Worked in Agile software development teams. Author of 17 publications, 2 patent applications.Sports analyticsSenior Analyst with the Milwaukee Brewers (MLB) for 2 seasons. Improved First-Year Player Draft model by assessing, cleaning, and feature engineering third-party data (TrackMan, Synergy), developing a position player defensive metric and pitch execution grade. Worked part-time with a sports betting start-up company. Built data processing pipelines with DonBest and Sportradar datasets in MongoDB. Developed models to predict the side and spread for NBA games.Leadership and teamwork skillsTeam lead for multiple projects, resulting in Department Innovator of the Year award. Recipient of Award in Leadership and Ethics. Worked effectively across global, multi-disciplinary teams. Delivered millions of dollars in value to multiple business partners. Mentored junior researchers and data scientists. Served 2 years as United Way of Hunterdon County Board Member.Independent data science projects• Pitch type thrown: Trained model to predict the pitch type (of five outcomes) expected based on situation – count, pitcher, etc. – using full-season data provided by MLB team. The most significant feature was the pitcher’s identity. Final model was 25% more accurate than baseline prediction.• Full season batting average: Trained model to predict 2018 season batting average for 309 players based on March/April 2018 data. Augmented training dataset with additional pre-2018 season data from fangraphs.com. The most significant features were hits in March/April 2018 (positive correlation) and HRs in prior seasons (negative correlation). Final model had 0.019 mean absolute error.• Time to next goal: Developed duration model for time between goals using Kaggle MLS dataset. Average time from opening whistle and first goal (33 min) was significantly longer than time between goals (20 min). Identified teams that are more likely to score / give up consecutive goals.
-
Senior Analyst - Baseball Research And DevelopmentMilwaukee Brewers Jan 2022 - Oct 2023 -
Research AssociateExxonmobil Jan 2015 - May 2021Annandale, NjEnergy Systems Quantitative Researcher- Defined data model and developed pipeline for storing chemical properties in relational database.- Reduced 80% of planned experiments by developing predictive model for chemical properties.- Innovative work resulted in 1 provisional patent application and 4 journal publications.Key skills: Data augmentation; Supervised machine learning; Building data processing pipelines in python; Optimization in AIMMS; Data analytics and visualization in Tableau; Version control using gitData Engineer / Team Lead- Saved Gulf Coast business partner $15M/year by developing pipeline to process and clean datasets.- Enabled near real-time monitoring of process by developing key performance indicator dashboards.- Mentored 4 junior data scientists.- Recognized with Department’s 2018 “Innovator of the Year” award.Key skills: Working in agile environment; Working remotely; Mentoring; Building data processing pipelines in SQL and python; Data analytics and visualization in Tableau; Version control using gitProgram Lead- Defined and executed strategy for new program to measure and model chemical mixture properties.- Led hiring committee for 4 positions; mentored 8 senior researchers.- Supervised work of 3 external partners (two universities, one software company).- Innovative work resulted in 1 provisional patent application and 10 journal publications.Key skills: Building consensus; Mentoring; IP strategy; Empowering team members to take ownershipMentoring Advocates- Co-designed web-based database of mentors to facilitate proteges' (of all experience levels) ability to find potential mentors.- Led several "speed mentoring" events, both in person and virtually, to broaden proteges' networks and create ice breaking event.Global Organization for the Advancement of Latinos- Co-organized site-wide I&D Day.- Installed OS on donated laptops for Hunterdon County residents in need.- Science fair judge for Hudson County STEM showcase. -
Senior ResearcherExxonmobil Jan 2013 - Dec 2014Annandale, Nj- Individual Contributor: Developed model of composition and compositional property models for heavy petroleum oil, enabling research programs that sought ~$300M savings. Project required merging ~6,000 historical data records with ~200,000 new data records self-consistently. Compositional property models were developed to utilize the ~206,000 data records as inputs for a predictive model. Computing infrastructure necessitated a data reconciliation protocol be developed to reduce the ~200,000 new data records to ~15,000 while maintaining the full model’s predictive capability.- Individual Contributor: Developed heavy petroleum oil thermal conversion model that utilized the ~21,000 reduced data records as input and predicted the change in each data record’s value for a set of reaction conditions (e.g. temperature, pressure, and time). Estimated model parameters utilizing historical datasets; identified parameters that were invariant to available datasets and designed / supervised new experiments to address this gap. Tuned model accurately predicted lab experimental results.- New Hire Professional Development Program: As site-wide Chair, led multiple initiatives to assist ~150 site-wide new hires with assimilating into the Corporate environment, mentoring / networking, and outreach / community service (e.g. United Way Day of Action).
-
Advanced ResearcherExxonmobil Jul 2011 - Dec 2012Annandale, Nj- Individual Contributor: Developed and validated detailed first-principles pressure swing adsorption process model in gPROMS software. Process model accurately predicted pressure, temperature, and species breakthrough curves as a function of feed pressure, volumetric flow rate, and composition for multi-component system for flow rates spanning three orders of magnitude. Insight from process model influenced project leadership to change adsorbent layer thickness significantly from design basis. Process model accuracy met requirements for project to proceed with commercial demonstration unit design.- Individual Contributor: Developed detailed first-principles temperature swing adsorption process models in gPROMS. Identified key performance metrics for evaluating different adsorbent bed architectures. Influenced project management team to pursue development of one bed architecture and halt further research on other two architectures under consideration.- New Hire Professional Development Program: As department-wide Mentoring Chair, initiated a “Speed Mentoring” event for ~25 department new hires to address gap identified in new hire evaluation of onboarding process.
-
Quantitative ModelerCompetitive Balance Analytics Mar 2020 - Sep 2020- Increased client’s return on investment by developing duration model for NBA line movements.- Applied multivariate statistical analysis for 5 seasons’ worth of Don Best irregularly spaced time series data to forecast ‘time to next event’ for totals and against the spread line movements.- Improved model’s accuracy by identifying and correcting data engineering error in data extracted from external database.- Subject matter expertise in NCAA enabled rapid identification of missing datasets.Key skills: Predictive modeling using Keras library; Building data processing pipelines with MongoDB
-
Graduate Research AssistantMassachusetts Institute Of Technology Jan 2006 - May 2011Cambridge, Ma- Contributed to the development and release of a Java-based chemistry software package, Reaction Mechanism Generator (RMG): Reduced the runtime by a factor of 2; extended the chemistry set (from carbon, hydrogen, and oxygen) to include silicon and sulfur; improved kinetic database’s rate coefficient estimates for molecules with alcohol functional groups.- Labeled 7,000 entries of an online thermochemical database with unique species identifiers, to facilitate RMG’s mining of the database for thermochemistry and kinetics parameters- Constructed code to solve a 2-D flame, and employed the solver to validate a reaction mechanism for three hexadiene isomers: S. Sharma, M.R. Harper, W.H. Green; Combustion and Flame 157 (7) 1331-1345 DOI: 10.1016/j.combustflame.2010.02.012- Built and validated a pressure-dependent reaction mechanism for the pyrolysis and combustion of 1-butanol against eight unique datasets: M.R. Harper, K.M. Van Geem. S.P. Pyl, G.B. Marin, W.H. Green; Combust. Flame 158 (1) 16-41 DOI: 10.1016/j.combustflame.2010.06.002- Extended the validated 1-butanol pyrolysis and oxidation model to include 2- and tert-butanol, allowing comparisons between primary, secondary, and tertiary alcohols: K.M. Van Geem, S.P. Pyl, G.B. Marin, M.R. Harper, W.H. Green; Industrial & Engineering Chemistry Research 49 (21) 10399-10420 DOI: 10.1021/ie1005349 -
Graduate Teaching AssistantMassachusetts Institute Of Technology Sep 2008 - Dec 2008Cambridge, Ma- Designed MATLAB-based problem sets involving linear algebra, optimization, differential equations, and fitting models to data
-
Chemical Engineering Practice ExternshipGeneral Mills Apr 2007 - May 2007Plymouth, Minnesota- Project Leader: Mapped the response surface for a cereal production unit operation, and identified operating parameters that resulted in a two-fold expansion in the product range- Quantified sugar crystallization kinetics at industrial conditions -
Chemical Engineering Practice ExternshipNovartis Feb 2007 - Mar 2007Basel Area, Switzerland- Project Leader: Made recommendations on equipment purchases and process scheduling to increase solid tablet production efficiency- Modeled the scheduling of a pharmaceutical bioproduction plant and performed throughput analysis and debottlenecking
Michael Harper Skills
Michael Harper Education Details
-
Chemical Engineering
-
Chemical Engineering Practice
-
Cumma Sum Laude
Frequently Asked Questions about Michael Harper
What is Michael Harper's role at the current company?
Michael Harper's current role is Sports Analytics.
What is Michael Harper's email address?
Michael Harper's email address is mi****@****ers.com
What schools did Michael Harper attend?
Michael Harper attended Massachusetts Institute Of Technology, Massachusetts Institute Of Technology, University Of California, Berkeley.
What skills is Michael Harper known for?
Michael Harper has skills like Python, Jmp, Seeq, Hive, Hadoop, Apache Spark, Matlab, Apl, Gproms, Java, Chemkin, Gaussian 03.
Not the Michael Harper you were looking for?
-
-
Michael Harper
Passionate About Building Stronger Brands To Create Richer Lives And A Better WorldSilver Spring, Md4summitmarketing.com, summitmarketing.com, summitmg.com, summitmg.com3 +177030XXXXX
-
2gmail.com, navy.mil -
Michael H.
Austin, Tx1sigepmail.com -
Michael Harper
Vienna, Va6pathway4.com, usi.net, bnx.com, coretego.com, fortisphere.com, zspace.com9 +170338XXXXX
Free Chrome Extension
Find emails, phones & company data instantly
Aero Online
Your AI prospecting assistant
Select data to include:
0 records × $0.02 per record
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial