P

Prathyush Maniyam

About

Detail

AI/ML Engineer | Data Scientist | GenAI, RAG, LLM Expert | Advanced Analytics | MS in Business Analytics & AI, UT Dallas
Mountain View, California, United States
Contact Prathyush regarding: 
work
Full-time jobs

Timeline

work
Job
school
Education
folder
Project

Résumé

Jobs verified_user 0% verified
  • JPMorganChase
    Data Scientist
    JPMorganChase
    Aug 2025 - Current (1 year)
  • Community Dreams Foundation
    Data Scientist
    Community Dreams Foundation
    Aug 2024 - Current (2 years)
    Built and deployed scalable GenAI pipelines (RAG, Hugging Face, AWS Bedrock, GCP, LangChain, FastAPI) automating 1,500+ personalized messages daily.
  • C
    Data Scientist
    Castus Info Solutions Pvt Ltd
    Dec 2021 - Jun 2023 (1 year 7 months)
    • Increased sales conversion rates by 20% by developing and deploying predictive models using Gradient Boosting, XGBoost, and Random Forest to identify high-potential buyers for push campaigns. • Reduced data processing time by 35% by industrializing ETL pipelines with Python scripts, utilizing Pandas, PySpark, and SQL for efficient data extraction, transformation, and loading. • Enhanced marketing ROI by 20% by developing customer propensity models using Logistic Regression, Gradient Boosting, and Neural Networks to identify high-value customer segments for personalized promotions. • Improved customer segmentation accuracy by 25% by applying supervised machine learning algorithms (Logistic Regression, Decision Trees, K-Means Clustering), e
  • KPMG
    Data Scientist
    KPMG
    Dec 2021 - Jun 2023 (1 year 7 months)
  • Tech Mahindra formerly Mahindra Satyam
    Data Scientist
    Tech Mahindra formerly Mahindra Satyam
    Jul 2017 - Dec 2021 (4 years 6 months)
    • Boosted stakeholder trust and model adoption by applying SHAP for explainable AI, improving model interpretability and transparency for executive audiences. • Reduced churn by 18% by optimizing churn prediction models using Support Vector Machines (SVM), driving retention-focused initiatives. • Enhanced demand forecasting accuracy by 22% for inventory management by implementing ARIMA, LSTM, and Prophet models, supporting operational planning and reducing stockouts. • Reduced machine downtime by 25% by applying predictive maintenance algorithms (XGBoost, Random Forest) to proactively identify and address machine failures. • Built forecasting models with 80% accuracy using ARIMA and LSTM to predict monthly order volume for refills, supporti
  • Dell Technologies
    Data Engineer
    Dell Technologies
    Jul 2017 - Dec 2021 (4 years 6 months)
  • BHEL Hyderabad
    Summer Intership
    BHEL Hyderabad
    May 2016
    As an intern at Bharat Heavy Electricals Limited (BHEL), I had the unique opportunity to delve into the fascinating world of electrical engineering. During my tenure, I embarked on a comprehensive study of the construction and operational intricacies of a Turbo Powered Electrical Generator in real-time.
Education verified_user 0% verified
  • The University of Texas at Dallas
    Master of Science - MS, Business Analytics
    The University of Texas at Dallas
    Jan 2023 - Current (3 years 7 months)
    Coursework: 1. Applied Machine Learning 2. Predictive Analytics for Data Science 3. Big Data 4. AdvanceStatistics for Data Science 5. Business Analytics with R 6. Database Foundations for Business Analytics 7. Deep Learning & Neural Networks 8. Generative AI and Large Language Models (LLMs)
  • GITAM University Hyderabad
    Bachelor of Technology - BTech, Electrical and Electronics Engineering
    GITAM University Hyderabad
    Jan 2013 - Jan 2017 (4 years 1 month)
Projects (professional or personal) verified_user 0% verified
  • C
    Credit Risk Analysis
    Developed an XGBoost model to predict credit default using the American Express dataset, maximizing revenue while controlling default rate. Conducted data processing, Hyperparameter Tuning, Bias/Variance Analysis, and SHAP analysis. Evaluated and demonstrated superior performance and interpretability compared to Neural Networks and Ensemble Models.
  • R
    Resume Chatbot
    Developed and deployed a scalable resume management application on GCP Cloud Run services with dual deployment modes, integrated Python Fast API backend, LLMs, and Llama Index framework with LangChain for efficient data extraction and query handling. Improving resume processing efficiency by 40% and query response time by 50%, while enhancing security using IAM authentication and improving user Experience.
This is a community-created genome.