Chaitra Tadaga Prabhukumara
Thousand Oaks,CA
Summary
Accomplished AI/ML Engineer with 6+ years of experience designing and deploying end-to-end Generative AI and Machine Learning solutions across pharmaceutical, fintech, and enterprise domains. Proven expertise building Custom GPTs with Actions, RAG systems, and LLM-powered analytics on Amazon Bedrock, SageMaker, Azure OpenAI, and Azure AI Foundry, alongside production ML models using XGBoost, Random Forest, LSTM, and BERT. Track record of measurable business impact - 15% churn reduction, 20% engagement lift, 85% model accuracy, and 14% query performance gains on migrated data platforms.
Overview
11
11
years of professional experience
Work History
Associate IS Business System Analyst
AMGEN
Thousand Oaks
10.2023 - Current
- Designed and deployed a Custom GPT with Actions integrated with the EDF enterprise database, enabling 50+ non-technical business users to query structured data in natural language and reducing average analyst turnaround time on ad-hoc data pulls by approximately 60%.
- Engineered GPT Action schemas (OpenAPI specifications) exposing 15+ curated database operations to the LLM, enforcing role-based access controls, row-level security, and audit logging for pharmaceutical compliance.
- Integrated the Custom GPT with internal analytics and reporting tools, automating dashboard generation and formatted output reports and eliminating ~20 hours/week of manual reporting effort across DQMS teams.
- Authored and iteratively refined prompt templates and system instructions to optimize response accuracy, consistency, and domain-specific tone across pharmaceutical data workflows.
- Performed data mapping between Trackwise and Veeva systems across multiple integration tracks, ensuring 100% data integrity, lineage, and traceability during DQMS integration initiatives.
- Gathered requirements, developed use cases, and performed system analysis to translate business problems into technical specifications.
- Skills: Veeva QMS, Open AI, Databrick, MySQL, AI Workbench
Data Engineer
VISA
Palo Alto
02.2022 - 07.2023
- Designed and implemented a real-time data pipeline using PySpark on distributed Spark clusters, processing semi-structured data from 5+ source systems and producing analytics-ready datasets for downstream product and risk teams.
- Led the migration from SQL to PostgreSQL using the Tusker migration tool, improving query performance by 14% and reducing average report generation latency across downstream analytics workloads.
- Configured spark-submit with speculative execution and tuned shuffle, memory, executor, and partition parameters, stabilizing 100+ production Spark jobs handling high-volume payment transaction data.
- Automated ingestion and transformation workflows using Python and Unix shell scripting, replacing manual processes and reducing operational toil by an estimated 30% for the data platform team.
- Tuned ETL component performance and optimized Teradata queries to consistently meet SLAs on high-volume batch processing.
- Built Tableau dashboards and pivot-table reports for business stakeholders, translating raw transaction data into actionable insights for product and risk teams.
- Wrote scripts and processes for data integration, data validation, and bug fixes across multiple production environments; authored Hive queries on large analytical datasets.
- Skills: Spark, Java, Scala, Python, JSON, SQL, Linux Shell Scripting, Parquet, Avro, AWS, Tableau
Data Scientist
Agile data Inc
Katy, TX
06.2021 - 12.2021
- Built predictive churn models using random forest and gradient boosting on 1M+ customer records, achieving 85% model accuracy and driving a 15% reduction in customer churn rate across target segments.
- Developed an LSTM and BERT deep-learning pipeline for topic classification on a large customer feedback dataset, benchmarked against classical models (SVC, Logistic Regression, Linear SVC, XGBoost, Logistic Regression CV) across accuracy, latency, and interpretability.
- Contributed to a recommendation system that drove a 20% increase in customer engagement through personalized content delivery.
- Conducted large-scale data cleaning, preprocessing, and feature engineering to ensure model input quality and maximize predictive power.
- Performed EDA to identify trends, patterns, and actionable signals for marketing and product teams.
- Created Tableau visualizations to present model outputs and insights to executive stakeholders, improving data-driven decision-making processes.
- Skills: Python, Tableau, Scikit-learn, Linear SVC, XGBoost, Logistic Regression, LSTM, BERT, TensorFlow, PyTorch
Data Scientist
SM Search System LLP
Bangalore, India
10.2015 - 12.2019
- Improved data quality & performance by 40% by analyzing, verifying and modifying SAS & Python scripts.
- Conducted statistical analysis on customer data to identify patterns and trends, leading to a 15% reduction in churn rate.
- Analyzed large-scale data sets using Python and SQL to extract actionable insights and identify trends.
- Used AWS EMR to transform and move large volumes of data across AWS data stores including Amazon S3 and DynamoDB, and designed partitioning / bucketing schemas enabling faster analytical query retrieval.
- Developed predictive models using machine learning algorithms to optimize marketing campaigns, resulting in a 20% increase in customer acquisition.
- Developed and maintained dashboards and reports using Tableau for real-time monitoring of key performance indicators.
- Analyzed high-volume log data using Splunk SPL to extract insights and support operational decision-making.
- Performed end-to-end architecture and implementation assessments across AWS services including EMR, EC2, Redshift, RDS, Lambda, and S3; configured AWS CLI auto-scaling and CloudWatch monitoring, alerting, and operational dashboards.
- Built Kibana dashboards on Logstash data and integrated source and target systems into Elasticsearch for near real-time log analysis and end-to-end transaction monitoring.
- Continuously monitored and managed Hadoop clusters through Cloudera Manager; worked with multiple file formats (Parquet, Avro, DAT, JSON) and compression codecs (Gzip).
- Implemented UNIX scripts to define use case workflows, process data files, and automate recurring jobs; delivered agreed user stories on time every sprint as part of a Scrum team.
- Skills: Python, Spark RDD, Spark SQL, AWS S3, EMR, Redshift, RDS, Lambda, Hadoop, Cloudera, Elasticsearch, Kibana, Logstash, UNIX
Education
Doctor of Business Administration -
Westcliff University
01.2028
Master of Science - Data Science
University of New Haven
01.2021
Bachelor of Engineering - Computer Science & Engineering
Visvesvaraya Technological University (VTU)
01.2015
Skills
- Generative AI & LLMs
- Custom GPTs
- GPT Actions
- OpenAPI Schemas
- Prompt Engineering
- RAG
- LangChain
- Vector Embeddings
- Foundation Models
- Claude
- Titan
- GPT-4
- Amazon Bedrock
- Bedrock Knowledge Bases
- Bedrock Guardrails
- Bedrock Agents
- Machine Learning
- XGBoost
- Gradient Boosting
- Random Forest
- Logistic Regression
- SVC
- Linear SVC
- Classification
- Regression
- Clustering
- Feature Engineering
- Hyperparameter Tuning
- Cross-Validation
- Deep Learning & NLP
- LSTM
- BERT
- Transformers
- Neural Networks
- Transfer Learning
- Text Classification
- Topic Modeling
- Sentiment Analysis
- Tokenization
- AWS AI/ML
- SageMaker
- SageMaker Studio
- SageMaker Pipelines
- Textract
- Comprehend
- Polly
- Transcribe
- Rekognition
- Lambda
- S3
- EMR
- Redshift
- RDS
- DynamoDB
- EC2
- CloudWatch
- IAM
- Data Engineering
- Apache Spark
- PySpark
- Spark SQL
- Spark Streaming
- Kafka
- Kubernetes
- Hadoop
- Hive
- Cloudera
- Parquet
- Avro
- JSON
- ETL
- Real-Time Pipelines
- Snowflake
- Teradata
- MLOps & DevOps
- Model Deployment
- Model Versioning
- CI/CD
- Git
- SVN
- Docker
- Workflow Automation
- Unix Shell Scripting
- Agile
- Scrum
- Programming
- Python
- SQL
- R
- Java
- Scala
- HTML
- CSS
- ML Libraries
- PyTorch
- TensorFlow
- Keras
- Scikit-learn
- Pandas
- NumPy
- Matplotlib
- Seaborn
- Databases
- PostgreSQL
- MySQL
- SQL Server
- MongoDB
- Elasticsearch
- Analytics & BI
- EDA
- Hypothesis Testing
- A/B Testing
- Time Series Analysis
- Tableau
- Kibana
- Power BI
- Logstash
Timeline
Associate IS Business System Analyst
AMGEN
10.2023 - Current
Data Engineer
VISA
02.2022 - 07.2023
Data Scientist
Agile data Inc
06.2021 - 12.2021
Data Scientist
SM Search System LLP
10.2015 - 12.2019
Doctor of Business Administration -
Westcliff University
Master of Science - Data Science
University of New Haven
Bachelor of Engineering - Computer Science & Engineering
Visvesvaraya Technological University (VTU)