SUMMARY:

• As a leader of Analytics, deep expertise overseeing technology initiatives aligning tactical solutions and strategic roadmaps supporting business goals.
• Strong reputation as being highly effective at technology selection, development and managing complex projects and cost - effective platform implementation with a vision for scalability, growth, and practicality.
• Partnered with leading technology, data providers and exchanges for cost-effective data solutions in region
• Versatile, intuitive and result-oriented data scientist with excellent integration of machine learning algorithms on statistical data.
• Perceptive interpersonal communication with business partners and teams.
• Having strong quantitative analytical skills and development experience.
• Excellent problem analyst producing creative and more practical solutions
• Substantial hands-on analytical experience with large datasets of text, image etc.
• Expertise includes abstracting and quantifying the computational aspects of the problems, designing and applying deep learning and AGI based algorithms
• Excellent knowledge in Azure compute services, Azure Web apps, Azure Data Factory & Storage, Azure Media & Content delivery, Azure Networking, Azure Hybrid Integration, and Azure Identity & Access Management. Experience in Automating, Configuring and Deploying Instances on Azure environments and in Data centers

TECHNIQUES:

Skills: Machine Learning, Deep Learning, Predictive Analytics, Applied Mathematics, and Computational Intelligence; Social Media Analytics, Reinforcement Learning, Robotic Process Automation, IOT Analytics, Text Mining, Sentiment Analysis, Artificial Intelligence, Natural Language Processing, Machine Vision and Computer Vision, OCR, TTS, NLG, Convolutional Neural Nets, Recurrent Neural Nets, Generative Adversarial Neural Nets, Spiking Neural Nets, RLS, QAMs, AGI, Hadoop, Yarn, Hive, HBase, Kafka, Flume, Storm, Spark, MLlib, Caffe, Theano, Tensor flow, Keras, PyTorch, MXNET

Technologies: Java, SAS, SPSS, Julia, R, Mat lab, Weka, XML, Scala, Python, JavaScript, IBM Watson, Databricks, Azure, H2o.ai, Data Robot, AWS, Google cloud platform, Hortonworks, Cloudera, RapidMiner, Alteryx, Qubole, Feature Labs

PROFESSIONAL EXPERIENCE:

Confidential

Director/Principal Data Scientist

Responsibilities:

• Worked with Confidential Financial services team as a director and developing solutions for different Artificial Intelligence use cases and building architectural road map to implement AI strategies for internal and external partners
• Focused on product optimization and defined research optimization road-map for maximizing revenue by customer engagement and responsible for the delivery of complex analytic engagements, infrastructure management, business information, infrastructure solutions, and strategy service
• Focused on making asset connectivity and management to the Industrial Internet of Things (IoT) increasingly seamless and intelligent.
• As the lead technical product manager for the team, I designed architectural road map for IoT fleet product management.
• My focus was on Industrial IoT (IIoT) specifically networking & big data platforms that securely connect millions of "critical things “developed & drove the execution of strategic vision for IoT Platform that provides big-data analytics & Asset Performance Management (APM) capabilities for IoT devices
• Worked on some small-scale solutions for clients: Fat fingering in trade analysis using Isolation forest and 1D Confidential ; Anti money laundering (alert routing & anomaly detection) using SVM AND CART; Multichannel attribution modeling based upon game theory and Markova chain; Insurance claim automation: Claim reporting using speech recognition, damage assessment using machine vision, customer sentiment using sentiment analysis, Document verification using OCR etc.

Confidential, Phoenix, AZ

Principal Data Scientist

Responsibilities:

• Designed and built real-time contextual behavioral personalization system using econometric and ML to predict behavior of the customer and help them to navigate through products of their choice. Led initiative to develop low latency models to learn and predict, enabling it to respond constantly to changes in customer’s payment behavior at various stages, with accuracy over 90%. Responsible for providing loyalty updates and future strategy road maps to senior management
• Used NLP techniques in developing a model deriving insights from Social Media postings. The model is used in the offering more customized card to customers to meet their expectations. It attempts to analyze trends from well performing cards by identifying intent and sentiment, agents use insights in customer acquisition. The models use Python, HAWQ and Open source including NLTK, Genism and Mallet. Developed a model that detects topics from historical and streaming chats
• Conducted assessment of insider threat email data using deep learning Convolutional Neural Networks; obtained over 90% correct classification of anomalous email obtained from open source CERT database. Latent Dirichlet Allocation used as the clustering technique, clustered all the columns into a set of discrete domains based on the similarity of the feature vectors. Developed language independent text summarization algorithm for the company's new core NLP engine

Confidential, NY

AVP, Advanced Analytics

Responsibilities:

• Designed and developed an algorithmic model in predicting likelihood of deals to reach certain sales milestones. The main objective is for sale teams to focus their attention on accounts with lower likelihood to convert. The model is developed in R using GBM with 7 features. The implementation of the model resulted to an annualized expected incremental lift by $2.32M. Designed models that deliver perspective insights to sales agents via an augmented Rafael
• Developed deep learning models for image classification using Caffe, CUDA, and cuDNN for our web extension which display customized contents in webpage based on customer p. Preprocessed images using OpenCV which included resizing, rotation and random crops to generate rich dataset. Realized an edge-cutting Subspace Clustering algorithm for high dimension sparse matrix, experimented for classification of MNIST image with accuracy of 98.14% and face image recognition
• Spearheaded an initiative for building new and additional features in the core product.
• Built a Journey Engine based on NLP, Text Cleaning, Text Mining and Aggregations. Built predictive models using XGB, DL( Confidential /Word2vec), Regularization & Ensemble Modeling for customer behavior predictions. Implemented recommendation Engines / Frequent Pattern Mining for customer needs.
• Performed Forecasting and Segmentation using Text Clustering Confidential, Graph mining and Page Rank analysis.
• Researched several topics in Natural Language Understanding, including word sense disambiguation and resolution as part of this project.
• Implemented named entity recognition ( Confidential ) methods on emails, feedbacks and messages to enable smart visuals in corporation in the videos.

Confidential, River woods, IL

Senior Data Scientist, Risk Management

Responsibilities:

• Responsible for the vision, strategy, architecture and engineering of data platforms including Time Series data, Real-time data, Big Data Eco-system, High Performance Storage & access design, Data Analytics, Business Intelligence, Master data management, and Metadata management. Lead the Enterprise Data Strategy Team to deliver data and analytics solutions to meet business needs in support of the enterprise defined business goals set by our senior management and executives
• Orchestrated an initiative which involves two hybrid models by combining two different neural network techniques for customer retention prediction, which are back-propagation artificial neural networks (ANN) and self-organizing maps (SOM). The hybrid models are ANN combined with ANN and SOM combined with ANN
• Designed and conducted experiments on data collection for reverse image search analysis using small sample size statistical analysis and Provided insight to team about results. Ported MATLAB face synthesis code into Python code, which required recreating a few built-in MATLAB functions including morphological filtering and NaN in-painting
• Built collaborative relationships with key individuals within lending groups to understand business priorities and explore the broader use of stress testing tools and analytics to add insights into portfolio risks and support business decisions
• Performed sentiment analysis on email feedbacks from customers to determine positivity using NLP and Text mining

Confidential, River woods IL

Senior Big Data Engineer

Responsibilities:

• Led team to design and build big data analytics integrated platform to enable businesses with OTIS, implemented risk management system that have imported trade positions and aggregated positions to provide portfolio level risk analytics
• Responsible for migrating DFS’s data transformation processes onto the Hadoop platform, which reduced data processing time by an average of 25%. Involved in expansion of mortgage loan profitability by $6 million by enhancing data mining capabilities and integrating acquired loan portfolio data into a Hadoop data operating system for better data processing

Confidential

Senior Software Engineer

Responsibilities:

• Designed various business models and channels (destination portal, SaaS model and CRM integration) for delivering the technology. Worked closely with domain experts to understand various challenges involved at different business systems
• Developed Raw material purchasing system and Order and Delivery management system, and Participated in the analysis and design of it.
• Actively participated in data modeling, data warehousing and complex database designing