Bachelor of Engineering (B.Eng.), Materials Science and Engineering @
I am excited to be in the middle of the birth of the Data Science community and its rapid advancement into many new domains. I like to tackle new problems and map them into mathematical models to be solved by leveraging the latest advancements in big data technology. Recently I became very interested in understanding the consumer
I am excited to be in the middle of the birth of the Data Science community and its rapid advancement into many new domains. I like to tackle new problems and map them into mathematical models to be solved by leveraging the latest advancements in big data technology. Recently I became very interested in understanding the consumer psychology by understanding their behavior using various data resources available.
At NativeX, I am using an arsenal of data science tools to derive insight from our extensive data trove. My current projects include feature engineering and clustering to deliver personalized targeted mobile advertisement. I am also involved in developing strategies for our cloud based experimental and production modeling pipelines.
During my PhD I have built up experience in software development of computational simulations and mathematical algorithms to solve engineering problems involving extremely large datasets and which require modern statistical inference techniques. I have a solid foundation in high performance computing and numerical techniques, and in their application to computational engineering and physics. I have extensive leadership abilities with a knack of bringing smart people together to solve problems. Also, I have excellent communication skills, and am frequently invited to deliver talks and lectures internally and externally.
Relevant skills and competencies include:
MapReduce (Java), Hadoop Streaming (Python), MRJobs, basic knowledge of Spark.
S3, EC2, EMR.
Python, SQL, C, UNIX Shell Scripts, AWK, Mathematica, familier with Java.
Statistics & Modeling:
Multivariate Linear and Non-Linear Regression, Neural Networks, Time Series and Sequential Analysis, Support Vector Machines, Clustering, Principle Component Analysis, Decision Trees, Recommender Systems, Molecular Dynamics, Monte Carlo, Phase-Field Type Models.
Senior Data Scientist @ From June 2015 to Present (7 months) Data Scientist @ • Feature Engineering - Designed and implemented over 100 behavioral and non-behavioral features using first and third party data to improve models’ predictions and increase revenue.
• Recommender system - Built recommender systems using collaborative filtering (NMF) to predict users’ app preferences based on their engagement to serve better targeted ads.
• ETL MapReduce (Java) - Created ETL pipeline using MapReduce (Java), MRjobs and Python Streaming to filter the raw Cassandra records and produced purposeful compressed data (Avro) ready for consumption by the modeling pipeline (TBs scale).
• Experimental pipeline - Designed and managed the cloud experimental pipeline to handle the big influx of features and hyper-parameter tuning to optimize the predictive models.
• A/B Testing - Executed A/B testing and deep logging analysis for live traffic to optimize revenue and to fine-tune the production pipeline.
• Production pipeline in the cloud (AWS) - Spearheaded the data science team in migration of the modeling pipeline to AWS in record time and reduce the production time by 75%.
• Ad hoc analysis - Employed MapReduce and Python(Pandas) to perform event level analysis to assist the executive team decisions on products.
• Leadership and management - Led a team of 5 on multiple projects including feature engineering, forecasting and deep learning with extremely tight deadlines to improve eCPMs.
• Teaching - Carried out multiple tutorials (internal and external) on various topics ranging from basic python to MapReduce in Java. From March 2014 to June 2015 (1 year 4 months) Graduate Trainee @ Enhanced performance of current solidification models by employing numerical optimization methods in MPI. Utilized various analytical and computational approaches to characterize the effects of impurity migration in hybrid atomic/continuum models. From May 2012 to November 2013 (1 year 7 months) Montreal, Canada AreaVisiting scholar @ Simulated over 100 years of computer time using molecular dynamics models to assess the effects of the high concentration limit of Cu-Ni system on solute trapping. Streamlined complex analysis code to take advantage of current multiprocessor techniques. From May 2010 to August 2010 (4 months) San Francisco Bay AreaVisiting Scholar @ Applied various algorithms to assimilate and analyze over 10TB of data. Investigated rapid solidification in dilute binary alloys to understand core physics of processes involved. From May 2009 to August 2009 (4 months) Sacramento, California AreaProcess Development Engineer/Supplier Quality Engineer (Intern) @ Devised best practices and solutions for robust manufacturing of high-end electronic products in collaboration with a six-member R&D team. Identified customer requirements and criteria to discover areas of improvement for quality, cost, reliability and delivery.
• Initiated thermal and mechanical test plans on silicon chips that improved reliability of electronic circuit boards. Conducted the collection and statistical analysis of new data to standardize testing of lead-free solder joint.
• Boosted customer satisfaction level as a result of implementing green and black six sigma strategies that minimized defects and reduced production time in manufacturing processes.
• Increased potential business leads by cultivating relationships with academic institutions seeking facts on emerging technologies.
• Improved risk management by formulating cost benefit analyses on behalf of customers to resolve product quality issues. From May 2006 to September 2007 (1 year 5 months) Toronto, Canada Area
Doctor of Philosophy (PhD), Computational Materials Science and Engineering @ McMaster University From 2008 to 2013 Bachelor of Engineering (B.Eng.), Materials Science and Engineering @ McMaster University From 2003 to 2008 Henry Humadi is skilled in: Data Science, Machine Learning, Python, Feature Engineering, Engineering, Physics, Data Analysis, Simulations, Algorithms, Big Data, Numerical Analysis, Mathematical Modeling, Matlab, MPI, Modeling
Looking for a different
Get an email address for anyone on LinkedIn with the ContactOut Chrome extension