Krishna Chamarthi

Lead development and deployment of production machine learning and Generative AI solutions supporting marketing analytics, operational workflows, and enterprise knowledge systems. • Designed and deployed production ensemble propensity models for MedSupp lead scoring, improving targeting effectiveness and contributing to a 25% increase in sales (AUC ~0.75). • Developed interactive Streamlit dashboards and scenario-based analytics to visualize model outputs, sales projections, and KPI impacts, enabling business teams to evaluate strategies and make data-driven decisions. • Built predictive models for insurance claims adjudication logic, translating complex legacy business rules into scalable ML-driven insights and improving analytical efficiency. • Implemented model monitoring and drift detection frameworks using PSI/CSI and Kolmogorov–Smirnov tests to ensure model reliability and performance over time. • Developed an internal LLM-powered knowledge assistant using RAG architecture (LangChain, OpenAI, FAISS) to support policy and eligibility queries across enterprise documents. • Led Dataiku platform adoption, including project migration, deployment workflows, and hands-on enablement to improve model deployment consistency and team productivity. • Operationalized ML workloads with infrastructure teams using AWS SageMaker, EMR, Docker, and CI/CD pipelines, enabling scalable and reliable model deployment. • Provided technical leadership and mentorship to data scientists through model reviews, best practices, and guidance on production ML development.

🔹 Passionate about leveraging the power of Machine Learning to drive business growth and innovation, I specialize in analyzing complex business needs and opportunities to develop cutting-edge solutions. 🔹 As a Data Scientist, I have successfully performed functional and technical analyses within client organizations and strategic customers, understanding their unique Machine Learning requirements. 🔹 Committed to rapid and iterative development, I have played a key role in creating validated minimum viable solutions that effectively address specific business challenges. 🔹 With a keen focus on performance optimization, I excel in feature engineering, harnessing the latest concepts of Reinforcement Learning to develop state-of-the-art algorithms for anomaly detection. 🔹 My expertise in Auto-encoders and Robust Auto-encoders has enabled me to achieve exceptional results, with a remarkable 98% recall rate in anomaly prediction. 🔹 Proficient in data analysis using MySQL and Spark SQL, I am adept at conducting in-depth studies and creatively utilizing new and existing data sources. 🔹 Additionally, I have successfully developed masked RCNN using ResNet50 weights for detecting telecommunication devices, employing both transverse and active learning to achieve superior recall performance. 🔹 Collaborating closely with product development teams and client partners, I have played an integral role in industrializing machine learning models and solutions, seamlessly integrating them into client offerings. 🔹 Committed to knowledge sharing and continuous learning, I actively engage with MI communities within the client organization, championing new technologies to drive innovation. 🔹 As part of my commitment to advancing the field, I actively engage with the external ecosystem, collaborating with academia, technology leaders, and open-source communities to shape the future of Machine Learning.

• Retrieves data using SQL within the Netezza framework and perform ETL operations. • Identifies data types (transactional, text, clickstream, etc.) and volume of data. • Handling large datasets and Perform data manipulations which is required for data modelling. • Optimizes code to make ongoing data retrieval as quick as possible. • Calculated LTV of the customers and LTV of customer by product wise. • Predicted the LTV by using GLM. • Used Pandas, Numpy, Seaborn, Scipy, Matplotlib, Scikit-Learn, NLTK in Python for developing various machine learning algorithms. • Perform A/B testing and validating the performance of the promotion and identify the potential areas of improvement. • Run Catalog Response model using decision tree, random forest and other ensemble models. • Scoring the model and compare with RFM to identify the importance of variables. • Built the multiple regression model for the identifying the keys variable for the business.

• Developed HR metrics on retention and recruitment • Worked closely with HR in developing retention and term dashboards • Worked closely with key HR units and business areas for developing additional reports which are part of internal projects which include comprehensive analysis of trends, KPI metrics, predictive analytics and other • Partnered with HRIS to research best delivery methods of reports, forms, documents, etc. and create samples for user approval • Created templates for recording the performance of Assist in compiling, calculating, analyzing and reconciling a variety of data, as directed. • Created Attrition models using deep learning techniques using python libraries like, Scikit-Learn, tensorflow with Machine learning like regression models like Random forests, Adaboost and Gradient Boosting to identify areas for improvement during hiring. • Provide analytical and reporting support to Human Resources. • Developed, Designed and Supported inter active Tableau dashboards. • Analyze and evaluate trends in recruiting and retention and offer feedback to Leadership in order to modify strategies accordingly. • Using Machine learning algorithms like regression, random forests, decision trees for creating Iteration models • Identifies and surfaces best practices from existing environments and areas of opportunities to increase profitability, reduce risks, and improve operational efficiencies.

Strong analytical skills that help evaluate a given situation, consider alternatives and find feasible solutions ETL operations using Advance SQL Proficient with the Statistical Concepts & Software like SAS, SPSS, R and Python and Techniques along with strong data interpretation abilities. Used multiple Python libraries Pandas, Numpy, Seaborn, Scipy, Matplotlib, Scikit-Learn, NLTK. Using multiple Advanced ensemble models like Random forests, Adaboost and Gradient Boosting. Perform analysis using SQL for the given data and draw accurate inferences in accordance with the objectives of the analysis. Capable of handling multiple tasks, taking initiatives, working in groups or as an individual Excellent logical thinking and problem-solving abilities Ability to accustom in multicultural environment with keen interest in learning new things

Perform analysis of the given data and draw accurate inferences, in accordance with the objectives. Importing data from various environments , compile it together in prescribed format Assist the organization in performing data compiling & mining required to evaluate the given data .Prepare analysis reports and be responsible to answer any queries, complaints or suggestion in this regard.Make use of different research methods, procedures, techniques along with statistical and analytic tools.