Irina Max

Proven results driven Technical Lead with Innovative Principal Data Scientist Hands-on skills and 20+ years in the data industry. Focus on building Statistical/ML/AI solutions to analyze, forecast and interpret specific meaning and provide actionable business insights for management, creating optimized strategies for success and improved business performance.

Summary:

• Technical Lead, mentor and creative hands-on Principal Data Scientist. Result driven problem solver with 9 years extensive experience and ability to build and lead Data Science teams to success, develop Strategy and direct Machine Learning projects using Big data technologies, employ statistical techniques with Big data initiatives and tools to drive major operational business decisions for success. Ability to lead through ambiguity and change.
• Master degree in Economics with solid Statistical background and Post-Master’s education in Data Science and ML at Stanford University. Technology-savvy and mathematically-equipped professional with experience in a diverse set of skills in Data Science including but not limited to methods such as Machine Learning, Deep Learning, Bayesian algorithm, Times series and Regressions and other statistical techniques.
• Expertise in complex data sets: product testing, parametric data, manufacturing, capital equipment. Selecting and configuring appropriate technologies, strategies and programming languages required to ensure successful business impact. Examine existing data and product designs to generate hypotheses and plans for high-impact research
• 5 course Deep Learning Specialization by Andrew Ng in Computer Vision- Natural Language Processing, Speech Recognition: CNN, LeNet, AlexNet, ResNet, RNN, Sequence model, LSTM, Gated Recurrent Unit, Autonomous driving, Accurate face recognition, Automatic reading of radiology images, Image recognition and Transferring. Hyperparameter tuning, Regularization, Optimization and other techniques.
• Effectively presented analysis results, explaining complex data, interactive dashboards and charts in a concise manner using RStudio/Server, Tableau, JMP, WEKA, MOA, including Plots, Data tables, Maps and Pivot charts in R.
• Expert in Marketing data analysis, Churn prevention, Customer Life Value designing experiments with prediction, Modeling Time to Reorder(Survival Analysis, Cox Proportional Hazard Model, Kaplan-Meier Analysis)
• Provide Customers Choice models for Product Marketing, Hierarchical models for commercial choice modeling, interpreting customers behavior, infographics, business decision support.
• Deployed 93% accurate Sales Opportunities Forecasting Model by Quasi Binomial Logistic Regression with individual Time of Event prediction, which improved company sales . Presented analytical data on a National Map with Statistical data by state.

• R, Python, RStudio/Server, RSpark, PySpark, SQL, Tableau, Presto, JSON, AWS, Redhat, Redshift, HPC Cluster, EC2, GCP, Kubernetes Cluster, IBMWatson, DataBricks, SPSS, SAS, Matlab, Excel, CSS, Docker, XML/HTML, JavaScript 
• Flex, Shiny, Flask, VM, .NET, H2O, MOA, WEKA, Atom, Git, Kernel, Unix, Linux, OSX, MS Office
• MySQL, PHP, Oracle, PostgreSQL, SQLite, NoSQL, JDBC, OLAP, PCA, ETL, Zeppelin, HDFS, Hive
• TensorFlow, Keras, TF Estimator, MXNET, CNN, ResNet, RNN, Neural Style Transfer, NLP, Parameters Tuning 

06/2022 - Present: Apple: Sr Principal Data Scientist /Assistant Manager Experimentations AdPlatform

• Supporting ML Experimentation AdPlatform Management and leading Data engineers team for projects of analyses and design experiments across Apple.com, Channel Digital, email marketing, social media and paid media. Proposals of ideas and Statistical techniques to predict Marketing Strategies and optimize Apple AdPlatform.
• Lead Project for Minimum Sample size determination for traffic A/B testing on Experiments AdPlatform based on the KPI metrics. Calculation the minimum amount of traffic required for !% KPI(RPO, TTR) change.
• Manage and support cross functional teams for Data problems detection, investigations and resolving.

• Manage Data Science and engineering team, support and work with stakeholders to assess demands and suggest data science methods and strategy that provide practical, value-added answers for the company.
• Leader of Fraud detection for invoice images Project based on the text extraction with Tesseract and Optical Character Recognition. Image preprocessing with text extraction. Spell Check for Images and PDF documents. 
• Hiring, mentoring and educational support for engineers and leadership.

• Leading 2 teams of engineers and data scientists in analytics and sales. Providing ideas and network solutions with technology-enabled services that help our customers obtain actionable insights, exchange mission-critical information, control costs, optimize nationwide revenue opportunities to increase cash flow and effectively navigate the shift to value-based healthcare. Collaborate with management, internal and external information technology teams on resolving data issues, improve data collection as well as mitigation plans to avoid errors in the future. 
• Optimize analytics to drive insights and strategic direction across analytical projects, marketing, sales, finance, medical network and related functional areas.
• Leading company dashboard forecasting projects. Effectively presented analysis results, explaining complex data, dashboards and charts in a concise manner to large technical and non technical audiences. 
• Developed and optimized Nationwide pipelines for Medical Network Volume Claim forecast, Multilevel Time Series Volume Claims daily and monthly prediction for company dashboard with 98% accuracy and for individual customers.
• Nationwide Payment Automation Analysis and Eligibility Time Series Vector AutoRegression model for Monthly prediction payment volume by Day with 98% accuracy using Lifecycle across scale technologies. 
• Multivariate state-space model VAR, SVAR, VARMA, TBATS, ARIMA with Fourier function for waves seasonality forecasting models, Granger Causality and Crosscorrelation testing and deployment for CHC dashboards.
• 93% accuracy Sales Forecasting Models with individual prediction of Time of Event for every pending open sales opportunity.
• Leading AI NLP projects for US customer segmentation based on the webscrabing and internet-scale data across numerous product and customer touch points.
• Supporting client/product teams including finding, pulling data, cleaning, and manipulating data to build data sets, preparing analysis, interpreting data and making strategic recommendations.
• Mentoring and advising company data engineering team members using Statistical and Scientific approach.

• Lead and partner with managers, data scientists, analysts, and engineers on the team and in aligned support functions to deploy, maintain, and scale models as integrated data products. Design and support experiments to quantify business value. Hiring and Coaching.
• Team leadership and individual contribution to a variety of projects and products focused on integrating ML and NLP in business processes. Supporting and expanding existing data products via experiment design. Data migration.
• Credit Risk anomaly research. Descriptive analysis. Fraud detection with improvements by scripting analysis on very high volumes of data at a commodity and parametric level. Classification and predictive Models.

• Directing ML project for product optimization and customer service Satisfaction. NLP , Sentiment Analysis of customer interaction with Apple support group.
• Customer segmentation, text mining for an extensive variety of email raw data with defined Sentiment score of customer satisfaction. ML project Analysis of Sentiment components with forecasting, visualization
• Manage Unlabeled text data analysis, nltk, Vader and Sentiment Analysis for determining polarity scores.
• ML/Deep Learning Trainer as Member and Volunteer Leader with Women Who Code Silicon Valley
• Presented several training sessions with workshops on Deep Learning architecture and AI (Python, Tensorflow, Keras).
• Developed and selected course content. Generated presentation outline and slides.
• Demonstrated concept implementation Forward and Backward propagation using python programs
• Provided students with a directed Deep learning python functions for programming learning exercise 
• Supporting students with a working python program that demonstrated FC Neural Network Models.
• Provide students with additional information for research and study outside of presentation.

• Manager of Analytical Legacy Pipelines for Data Science Infrastructure support. Cross-collaborating with Infrastructure engineering and other teams, drove strategic initiatives for better data collection and reporting, ensured data integrity across multiple data sources, and reduced analysis time through automation and creative solutions. Infrastructure Data Science team support. Mentor to Resource to Production pipelines. Data modeling and visualizations, developing, monitoring and debugging legacy pipelines and dashboards.
• Automate analyses and build analytics data pipelines via SQL and python based ETL framework.
• Partner with cross-functional teams to identify directions and new opportunities requiring the use of modern analytical and modeling techniques.

• Developed Machine learning models for endurance of high quality Apple memory, BICS3 reliability, die sort parameters tuning using LASSO, XGBoost, Ranger, Random Forest. Ensemble models learning and experiments.
• Mentoring, providing technical and thought leadership on designing, prototyping, implementing and automating complex analyses of hardware data.
• Deep analysis using ML techniques of the HTPD/RTPD/LTPD tests data to define a model of FBC growth rate across the temperature for finding signals for failing memory bits growth.
• Memory Technology, experimenting with statistical tests and analysis for selecting and optimizing elements of memory. 
• ML Moving Average models for projection pre-production SLC, MLC, TLC single and multi die packages ICC memory characteristic 6s with and without PPM. Monte Carlo simulation presenting Six Sigma probability to understand the impact of risk and uncertainty in ICC memory forecasting models.

• Analytical and Data Science consulting projects for small business owners including: Financial Operations, Marketing, Sales, Quality, New Product Development, Customer Service, Supplier Quality, Global Supply Managers, Suppliers, Contract Manufacturers, Repair, Business Intelligence.
• Managed and maintained business database and structure, creating Business analysis with forecasting, visualization for Business strategy, risk management and creative decision support.
• Analyzed and managed marketing data to maximize pricing margins using SQL and R. Effective communication insights and recommendations to management in support of strategic decision-making. CLV and churn prediction using statistical survival analysis technique.
• Research and SWOT analysis for strategy, developing and best business decisions.

• Lead and organized financial database for a 25 million dollar American aid program. Implemented and analyzed financial models to distribute funds. Statistical Analysis Time-series data from over 50,000 water users examining water use, crop yields, equity of water distribution, and farmer satisfaction. Cross-functional supervised work with models of 3 teams at Kirgistan, Uzbekistan and Tajikistan.
• Management support by tracked distribution of funds to cooperating stakeholders. Time series.

• Manage and lead ETL and security for 50+ disparate databases. Administered, designed and implemented SQL custom database for the national pharmaceutical center to track the receipt and distribution of medicines .
• Managed inputs and updates to medical databases; debug designed transactions; planned database backups and restores; maintained data storage; managed and trained employees.

• Lead 100+ programming and financial projects, verified account balances, generated electronic signatures, performed encryption and data security tasks. Automated transactions robot development. Increased transaction speed by implementing a new security approach. Trained staff.Credit Risk and Payment Analysis.

MS and BS in Economics, Tashkent State University of Economics, GPA 3.9

Post Master’s Education: Machine Learning, Stanford University