Louden Motina

Experience

• 

  CompSource Mutual Insurance Co

  Feb 2015 - Jun 2017

  Data Scientist

  Worked within the Enterprise Applications Team as a Data Scientist. Project responsibilities include multivariate analysis to test the impact of the WorkSafe Champions Safety Program on policies. I also worked with the SIU (special investigations unit) to indicate the feasibility of detecting insurance fraud with Machine Learning and help gain insight• Operated within a Cloudera Hadoop environment, harnessing Python, SQL, and Tableau to unlock insights from extensive datasets• Conducted research to evaluate fraud predictive analytics scenarios, identifying patterns and predicting outcomes for new claims. This involved the development of predictive models and comprehensive data analysis• Employed a robust toolkit, including Python, Pandas, NumPy, and SciPy, for exploratory data analysis, data wrangling, and feature engineering• Evaluated Anomaly Detection Models, including Expectation Maximization, Elliptical Envelopes, and Isolation Forests, to enhance fraud detection capabilities• Extracted and harnessed data residing within the Hadoop Distributed File System (HDFS) on Cloudera for analysis• Crafted a comprehensive Tableau Dashboard, providing a powerful tool for presenting the organization's Annual Report• Explored the application of kernel density estimation in lower-dimensional spaces as a predictive feature in fraud detection• Conducted in-depth multivariate analysis on safety programs spanning multiple years, aiming to uncover valuable insights• Leveraged regression analysis to establish correlations between participation in safety programs and claims outcomes• Executed Hypothesis testing and rigorous statistical analysis to pinpoint statistically significant changes in claims following participation in safety programs• Utilized Tableau and TabPy to visualize complex analyses and present findings effectively• Managed the integration of fraud data with claims data, effectively handling large datasets with multiple observations Show less

• 

  McKinsey & Company

  Jun 2017 - Sept 2019

  Data Scientist & ML engineer

  With an enterprise resource planning solution, worked on a project where organizations can use data and analytics to better predict supply needs and anticipate disruptions. Worked with diverse and complex datasets also contribute to tangible business improvements by leveraging data science methods and advanced analytics.• Utilized machine learning and statistical modeling techniques to develop and assess algorithms aimed at optimizing performance, data quality, and accuracy within the industry's Supply Chain management• Managed version control through Git and implemented Jenkins for CI/CD pipelines to streamline code management and deployment• Established a playbook for events and classification containers, improving data management and classification processes• Took charge of data analysis, cleaning, and debugging efforts, ensuring the readiness of datasets and codebases for Quality Assurance.• Spearheaded the development of a proof of concept for a chatbot, showcasing its potential utility• Collaborated closely with product and program managers to define data analytics problems and formulate effective solutions• Designed and developed an application using Django, successfully moving it to deployment• Automated model training and deployment through ML operations, enhancing efficiency and scalability• Employed statistical analysis and data visualization techniques to uncover valuable patterns, trends, and insights for informed decision-making• Utilized predictive modeling to forecast outcomes and identify potential health risks for patients, facilitating personalized treatment plans and improved care• Leveraged statistical analysis and machine learning algorithms to identify target patient populations, refine trial designs, and assess potential risks• Ensured organizations' compliance with regulatory requirements, including HIPAA, GDPR, and FDA regulations, safeguarding data privacy and integrity Show less

• 

  Jefferies Financial Group

  Sept 2019 - Jul 2021

  Senior Data Scientist

  The Data Strategy project is a global initiative aimed at providing critical support to the Equity Research Department and its analysts. Our primary objective is to seamlessly integrate alternative data sources into their research reports. This integration combines the fundamental domain expertise of the research team with our technical skills to facilitate actionable analyses. We also maintain proactive collaboration with third-party data vendors to enhance their products. Our team's core focus revolves around the methodical, defensible, and actionable processing of extensive datasets, enabling us to construct narratives about real-world behavior and dynamics. Our work is founded on statistics, data engineering, and data science, with occasional forays into machine learning and artificial intelligence as tools to augment our analysts' capabilities.• Spearheaded the integration of diverse data sources, including alternative datasets, into the Equity Research Department's analytical framework, ensuring data quality, consistency, and relevance, in this job i collaborated closely with equity analysts to harness their domain expertise and translate it into actionable insights through advanced analytics techniques, such as regression analysis, clustering, and time series forecasting.• Applied robust data preprocessing techniques, including data cleaning, normalization, and feature engineering, to large and complex datasets, enhancing the accuracy and reliability of analytical outcomes using Pandas, NumPy), Data Cleaning and Feature Engineering• Leveraged statistical methodologies, such as hypothesis testing, ANOVA, and regression modeling, to extract actionable information from datasets and support hypothesis-driven research.• Applied machine learning and artificial intelligence algorithms, to augment research processes and empower analysts with predictive and prescriptive analytics with ML Framework such as Scikit-Learn & TensorFlow Show less

• 

  ABIK HEALTHCARE SERVICES, INC.

  Jul 2021 - Jan 2024

  Senior Data Scientist

  This data science project revolves around the collection and analysis of a company's diverse healthcare data stored in Amazon S3. The primary objective is to establish a robust data pipeline, leveraging Snowflake, to construct a comprehensive data warehouse and build a text summarization model and other machine learning models for business analytics. the snowflake warehouse helps in querying electronic health data derived from various sources, including patient records, clinical information, insurance data, billing records, and hospital transactions.•This project involved building a text summarization UI using LLMS , we use pretrained models from hugginface and fine tune the models using loRA and build a ui using langchain framework with streamlite on aws sagemaker .• Applied various generative AI techniques using the transformer architecture and nlp for text cleaning ,name-entity recognition, chatbots etc.• Worked on vector database like pinecone in building a RAG for easy retrieval of proprietary healthcare data we used ranking techniques on the vector database and best similarity search techniques to improve retrieval.• Build a rag using pandas and langchain framework ,pinecone and streamlite to expose an api for online inference of the model after finetuning. • Gather electronic health data from multiple sources, centralizing it in Amazon S3 for subsequent processing.• Apply data cleaning and transformation techniques to ensure data accuracy and consistency. Python, with libraries such as Pandas and NumPy, is utilized for this purpose.• Use aws databrew and glue for etl and use apache airflow for workflow orchestration and realtime from s3 to snowflakes• Build and maintain a Snowflake data warehouse to store and organize the cleaned data efficiently.• Perform SQL queries on the data warehouse to extract valuable insights and generate actionable reports.• Use pyspark and gpu based machines to build a text summarization model using Amazon sagemakeR Show less

• 

  Commonwealth of Kentucky Cabinet for Health and Family Services

  Jan 2024 - now

  Data Scientist

  Working on Medicaid data for data mining purposes and machine learning. Perform statistical analysis using python and sql