Manikanta Pudoka
I am a data engineer and data scientist with 3+ years of experience across healthcare, finance, and industrial domains.
About
I am a data engineer and data scientist with 3+ years of experience across healthcare, finance, and industrial domains. I build scalable ETL pipelines and analytics solutions using Python, PySpark, Apache Spark, Databricks, Airflow, dbt, and cloud platforms such as AWS and Azure.
I thrive in cross-functional teams, and I have guided a 4-member group to align roadmaps and accelerate releases. My work includes predictive modeling with scikit-learn, NLP-based feature extraction from EHR data, and delivering dashboards and analytics that empower decision-making in clinical and business contexts.
- Joined October 2025
Industry experience
- Healthcare
- Financial Services
- Manufacturing
- Education
- Software & Internet
More experts in St. Louis, United States
Work experience· 4 roles
Engineered Python-based pipelines to ingest, clean, and transform over 5 million rows of multi-source healthcare data, boosting downstream analytics throughput by 35 hours per month. Developed predictive risk models using scikit-learn and SQL, achieving AUC above 0.85 on test sets and enabling classroom case studies based on realistic risk stratification. Explored EHR datasets with SQL and NLP methods (tokenization, entity extraction) to derive ~10,000 clinical features for faculty research and student presentations.
Automated ETL pipelines using Python, SQL, and Snowflake to process over 1.2 million daily records in under 4 hours, enabling faster data availability for analytics teams. Scaled batch processing by integrating Databricks with Apache Spark to handle more than 20 million records per job. Deployed machine learning model endpoints on AWS SageMaker delivering approximately 150,000 transaction predictions per day. Architected data warehouse solutions on Snowflake and Azure Data Lake ingesting data from 7 business units, improving report generation time by 3 days per month. Validated compliance data in SQL and MongoDB by reconciling 500,000 regulatory records per sprint. Mentored a team of 4 junior engineers, shortening average delivery timelines by 14 days across projects.
Processed over 25,000 images using Python and OpenCV to build an object detection pipeline for image-based tasks. Streamlined feature extraction workflows via NumPy and Scikit-Learn to reduce model training iteration time by 30 hours across three prototypes. Prepared datasets of more than 1.2 million records using SQL and Pandas, ensuring consistency and enabling downstream analytics reuse.
Education
Hire Manikanta Pudoka today
Post a job and invite Manikanta, or contact them directly. No hidden commission.




