Nishikanta Mohanty

A highly accomplished Data Science professional boasting over 18 years of distinguished performance in designing data-driven solutions for pivotal projects across various sectors. My expertise spans Machine Learning, Quantum Computing, and SAS, with a current academic focus on earning a PhD in Quantum Computing and a Postgraduate degree in AI & ML. I have a proven track record in deploying algorithmic techniques for data processing to unravel analytically complex business challenges, turning vast amounts of raw data into actionable insights. My practical experience includes developing AI/ML solutions for sectors such as fraud detection, workforce analytics, and telecom, specifically in churn and propensity modeling. I excel in comprehending the full spectrum of business issues, then architecting and implementing scalable AI & ML and decisioning solutions. These solutions leverage statistical techniques and incorporate the latest in cutting-edge technology to address and solve these issues effectively. I possess hands-on coding skills in Python and SAS, allowing me to extract key insights from business data. Moreover, I am proficient in utilizing open-source frameworks such as scikit-learn and TensorFlow to craft advanced AI/ML solutions. My deep understanding of various techniques, including but not limited to linear regression, logistic regression, cluster analysis, decision trees, neural networks, recommendation systems, quantum optimization, and text mining, enables me to contribute significantly to data science initiatives, delivering results that drive business success and innovation.

• Star Performer Award – MEA (Tech Mahindra) 2011. Achievement: Successfully completed analytics project for MTN Nigeria resulting decrease of 5% churn and increasing 2% upsell.
• Best Technical Support Consultant – India and South Asia (SAS Institute India) 2013. Achievement: Successfully solving customer issues and ensuing renewals of 7.5Mn INR for a Large Insurance Client.
• Outstanding Performance Award – APAC (Tech Mahindra) 2017. Achievement: Successfully optimizing Optus Campaign System, ensuring error-free operations for iPhone lunches.

Project: Customer Fraud Detection | Client: Largest Australian Telco |Year: 2022- Mar- Current

• Collaborated with client fraud analytics team to analyse fraudulent behaviors, segmented transactional data, and developed classification models for early fraud identification across retail, telesales, and online channels.
• Used advanced techniques such as SMOTE and Clustered down-sampling to reduce bias and improve overall accuracy.
• Developed post-deployment monitoring, model evaluation, and model explainability reports.
• Played a vital role in customer decision-making by developing profitability matrices that aid in selecting and deploying fraud models.
• § Established Best Practices in Model Evaluation and Monitoring and added explainability capability

Project: Workforce analytics | Client: Australian Defence Force | Year: 2022-Nov – 2023-Jun

• Worked closely with ADF data teams to understand human resources and recruitment data.
• Established a workflow for summarizing Job and candidate data.
• Developed a two-stage approach utilizing clustering and classification methods to arrive at recruitment recommendations.
• Presented the findings to the ADF data team and explained the usage aspect.

Project: HR analytics | Client: Leading Australian Health Insurance| Year: 2023-Apr - 2023-Oct

• Worked Closely with client team to understand employee data.
• Established a workflow for summarizing healthcare workforce Data.
• Developed a Decision Tree Model to predict employee churn with emphasis on variable importance andxplainability
• Developed a two-stage approach for reducing employee churn using machine learning and decisioning.

2005-02 - 2007-07

Business Intelligence Developer

Satyam Computer Services , India

Selected Projects AI and ML

Project: Customer Fraud Detection | Client: Largest Australian Telco |Year: 2022-Mar- Current

§ Collaborated with client fraud analytics team to analyse fraudulent behaviors, segmented transactional data, and developed classification models for early fraud identification across retail, telesales, and online channels.

§ Used advanced techniques such as SMOTE and Clustered down-sampling to reduce bias and improve overall accuracy.

§ Developed post-deployment monitoring, model evaluation, and model explainability reports.

§ Played a vital role in customer decision-making by developing profitability matrices that aid in selecting and deploying fraud models.

§ Established Best Practices in Model Evaluation and Monitoring and added explainability capability

Project: Workforce analytics | Client: Australian Defence Force | Year: 2022-Nov – 2023-Jun

§ Worked Closely with ADF data teams to understand human resources and recruitment data.

§ Established a workflow of summarizing Job and candidate Data.

§ Developed a two-stage approach utilizing clustering and classification methods to arrive at recruitment recommendations.

§ Presented the findings to the ADF data team and explained the usage aspect.

§ Added explainability and monitoring features.

Project: HR analytics | Client: Leading Australian Health Insurance| Year: 2023-Apr - 2023-Oct

• Worked Closely with client team to understand employee data.
• Established a workflow for summarizing healthcare workforce Data.
• Developed a Decision Tree Model to predict employee churn with emphasis on variable importance and explainability.
• Developed a two-stage approach for reducing employee churn using machine learning and decisioning.

POC: Accident fatality Minimization | Client: An Australian Government Health Department | Year: 2021-Mar - 2021-Jun

• Worked Closely with Various Data Systems teams of the Health Department to Gather Data related to Accidents.
• Established a workflow of summarizing and cleaning data for various reports related to accidents.
• Created a Clustering model for identifying recurring accident hotspots and ambulance positions to minimize accident fatality.
• Prepared presentations and demonstrated the prototype to the non-tech audience.

POC: Credit line Risk Prediction | Client: A leading Australian Gaming and Entertainment Club | Year: 2021-Mar - 2021-Jun

• Worked Closely with Various Data Systems teams of the Club to Gather Data related to Patrons and Credit History
• Established a workflow of summarizing and cleaning data of borrowing patterns for Patrons.
• Created ensemble models using Random Forest, XGBOOST, AdaBoost, and Bagging methods for predicting the high-risk borrowers.
• Presented and implemented AutoML Features of SAS Viya with SAS Visual Analytics in the form of Analytics Dashboards

Project: Analytics Migration | Client: SingTel | Role: Technical Architect | Platform: SAS Base, SAS Eminer, SAS Enterprise Guide, Unix | Team Size: 5 | Year: Mar’14 – Jun’15

• Contributed significantly in re-designing existing statistical procedures and models as per procedures and models new data structure and data requirements of the new data warehouse.
• Involved in preparing solution documents, automating E-Miner models and design framework to accommodate the existing model functionality.
• Redesigned the Existing models of prepaid/postpaid Churn, Mobile TV, Home Broadband, and Entertainment bundles to leverage new sources.
• Oversaw regression testing of Model flows and re-tweaked models to achieve the desired accuracy.

Project: Telecom Analytics | Client: MTN Nigeria | Role: Model Developer | Platform: Oracle 10g, SAS9, SAS Enterprise Miner, SAS E Guide, Solaris | Team Size: 4 | Year: Jul’10 – Jan’12

• Architected solution for the marketing team comprising models for Churn, CLV, and Propensity by adding data-mining Capabilities to MTN Nigeria Bi stack, resulting in improved intelligence reporting.
• Worked closely with data warehouse teams to understand the various data sources for analytical models.
• Created models for customer churn, Cross-sell/Upsell using logistic regression, KNN, and Decision tree.
• Created Models for Customer Lifetime Value
• Created Control-M flows for batch scheduling models and associated ETL flows.

Project: Analytical Decisioning | Client: Leading health insurance provider | Year: 2022-Jan – 2022-Sept

• Collaborated with the customer data team to understand the customer's claim processing process and claim data.
• Developed a workflow for cleaning and organizing claim data.
• Developed decision flows Using the SAS Intelligent Decision platform for processing claims.
• Developed and embedded fraud classification models to claim decision flows.
• Worked with various teams to deploy the decision flows to containers.
• Established Decision Ops Workflow process using Jenkins and GitHub

Project: Campaign Engine | Client: Optus Australia | Year: 2015-Nov – 2018-Aug

• Provided supervision to SAS, PEGA, and Abinitio support and development teams for architecting technical solutions for the campaign platform, including new campaigns, SAS operations management, designing HLD, and troubleshooting.
• Designed and Architected Data Solutions for Campaigning and Warehouse Systems.
• Provided L2 and L3 support for the Campaigning team to solve critical issues.
• Held Operational Oversight on Business-critical systems.

Project: Telecom Datawarehouse | Client: one of the largest Middle Eastern Telecom | Platform: SAS EBI 9.2, IBM Data

Stage, HP UX| Team Size: 15 | Project duration: Oct’08 – Jul’10

• Worked closely with BSS, and OSS teams to understand the various data sources for Data warehouse.
• Created ETL workflows for loading Prepaid/Postpaid and CDR data.
• Developed analytical Reports for Marketing, Sales, and Operations
• Created Control-M flows for batch scheduling ETL flows.

Project: Consumer Datawarehouse | Client: GeMoney | Platform: SAS , Abinitio, HP UX| Team Size: 15 | Project duration: Oct’05 – Jul’07

• Created ETL workflows for loading credit card application data and transaction data.
• Scheduled ETL flows using Autosys.
• Solved Level 1 issues.

• Regression
• Decision Trees
• KNN
• Naive Bayes
• SVM
• Clustering
• Random Forests
• XgBoost, Adaboost
• PCA
• Recommendation engines, SVD
• Model Tuning, Pipelining
• Neural Networks (ANN, CNN)

• Marketing Automation
• SAS Intelligent Decisioning

• Quantum Algorithms
• VQE, QAOA, SVM
• Qiskit

• NumPy
• Pandas
• Scikit-learn
• SciPy
• XGBoost
• Seaborne
• Surprise
• TensorFlow

• Azure
• AWS

• Visual Analytics
• Visual Data Mining and Machine Learning
• SAS Studio
• Intelligent Decisioning
• Workflows
• Enterprise Guide
• SAS Visual Investigator
• SAS 9.4 Administration

• Linux
• Oracle
• Pl/SQL
• Shell Scripting

• Analysis of The Vehicle Routing Problem Solved via Hybrid Quantum Algorithms in Presence of Noisy Channels, 2022, https://ieeexplore.ieee.org/document/10214310
• Solving The Vehicle Routing Problem via Quantum Support vector machine, 2023, https://arxiv.org/pdf/2308.04849.pdf