Ml Engineer Deep Learning

M V S ANIRUDH, MAGANTI

317-***-**** **************@*****.*** LinkedIn GitHub Portfolio

PROFESSIONAL SUMMARY

AI/ML Engineer with expertise in building NLP and deep learning models, including metaphor detection (DistilBERT) and crop row segmentation (U-Net, 85% accuracy). Proficient in deploying scalable ML pipelines on AWS and GCP, with a proven track record of improving business outcomes (e.g., 20% boost in churn prediction accuracy at ADP). Passionate about leveraging Python, TensorFlow, and predictive modeling to solve complex problems in dynamic industries. SKILL SET

Programming: Python (NumPy, Pandas, Scikit-learn, Matplotlib, Seaborn, NLTK), SQL, R, Linux AI/ML: Linear Regression, Logistic Regression, K-Means, Gradient Boosting, NLP (Bert, DistilBERT, RAG, LangChain, LLM, Llama, FAISS), Deep Learning (CNN, U-Net, RNN, LSTM, GRU), Statistics (Hypothesis Testing, Anova, PCA, Bayesian Analysis, Time Series)

Tools/Frameworks: TensorFlow, PyTorch, Keras, AWS (Lambda, SageMaker, EK), GCP (Big Query), Apache Spark, Apache Airflow

Deployment and MLOps: Docker, Kubernetes, Kubeflow, MLflow, Flask, Fast API, Git, GitHub, GitHub Actions, Jenkins Data Visualization: Power BI, Tableau, Matplotlib, Seaborn EXPERIENCE

ADP

AI/ML Engineer. Dec 2023 – Aug 2024

• Developed NLP models for HR virtual assistant using BERT and LangChain, improving query resolution time by 30% through entity recognition and intent classification.

• Built predictive models for workforce management, achieving 20% higher accuracy in employee churn predictions using scikit- learn and Gradient Boosting.

• Designed real-time ETL pipelines with Apache Spark and Kafka, enabling instant insights into HR and payroll metrics for enterprise clients.

• Implemented ARIMA and Prophet models to forecast payroll volumes, reducing processing lag times by 15% and optimizing resource allocation.

• Created anomaly detection models with isolation forests and autoencoders, boosting payroll fraud detection accuracy to 95% and cutting investigation time by 40%.

• Automated payroll and HR reporting with Natural Language Generation, halving report generation time and delivering actionable insights.

• Deployed models using Docker and Jenkins CI/CD pipelines, reducing deployment time by 40% with robust version control.

• Utilized SHAP for explainable AI in payroll decisions, ensuring transparency and compliance with ethical AI standards. Taylor Corporation

AL/ML Engineer Feb 2023 - Nov 2023

• Developed a content-based recommender system using NLP techniques (SpaCy, Gensim, RAG), increasing user engagement by 20% through personalized content delivery.

• Built end-to-end ML pipelines with PySpark, MLflow, and Flask APIs, streamlining model deployment by 25% and enabling rapid iteration.

• Automated AWS Lambda functions and Snowflake ETL jobs for Adobe Analytics, improving data processing efficiency by 15% for marketing insights.

• Created a real-time chatbot with Azure Bot Service and LangChain, boosting response accuracy by 30% via advanced NLU capabilities.

• Designed Power BI dashboards to monitor ML model performance, leveraging Apache Airflow for efficient batch data processing.

• Managed the ML lifecycle with Kubeflow, enabling scalable model training and experimentation for product teams.

• Applied PCA and hyperparameter tuning to optimize model accuracy, improving predictive performance by 10% for recommendation systems.

• Deployed models using Docker and GitHub Actions, ensuring reproducible workflows and minimizing deployment errors.

• Conducted data preprocessing with Pandas and NumPy, cleaning and transforming user data to enhance model reliability by 15%.

• Collaborated with product teams in Agile sprints to align ML solutions with business goals, improving project delivery timelines by 20%.

Hyundai Motor India Engineering Pty Ltd

Data Scientist, Jul 2019– Dec 2022

• Built multivariate time-series models with Python and SQL to forecast vehicle demand, improving production planning accuracy by 13%.

• Developed automated analytics pipeline using Python (Pandas), SSMS, and Power BI, saving 10+ hours/week in KPI reporting.

• Applied RFM segmentation and clustering with scikit-learn, boosting service retention campaigns by 5% through targeted insights.

• Conducted statistical audits on ADAS sensor data using NumPy and PCA, improving feature integrity by 30% for downstream ML models.

• Deployed models using Docker and GitHub Actions, enhancing reproducibility and streamlining deployment workflows. PROJECTS AND ACADEMIC ACHIEVEMENTS

Metaphor Detection (NLP): Developed a DistilBERT-based model to detect metaphors in text, achieving a state-of-the-art NLU loss of 0.045 using Python and TensorFlow.

Crop row and Leaf Segmentation (Deep Learning): Built U-Net models achieving 85% accuracy in crop row detection and 97% in leaf segmentation, enabling precise yield assessment and optimized planting. California Housing Price Prediction (Machine Learning): Developed and deployed an XGBoost Regressor model for California housing price prediction on AWS EC2, utilizing systemd for process management and Nginx as a reverse proxy to ensure efficient and scalable application performance. Credit card Risk Estimation (Data mining): Designed a fraud detection model using Random Forest and scikit-learn, reducing false positives by 15% on historical credit card data. Click-Through Rate Prediction Marketing Campaign Optimization: Developed a click-through rate prediction model on highly imbalanced marketing data (~1.6% positive rate), optimizing for log loss and calibrated probabilities rather than raw classification accuracy, ensuring reliable ranking of high-intent users. Bayesian Health Insurance Cost Prediction: Developed a Bayesian model using MCMC for predicting health insurance costs; achieved R-squared of 0.89 and RMSE of 0.46 using Python and R. Other Projects: Financial and HR Attrition Dashboards (Power BI, Tableau), Sales Dashboard (Power BI):, Credit Card Financial Dashboard (Power BI and SQL Server):

EDUCATION

Purdue University, Master of Science Jan 2023 – Dec 2024 USA

• Major : Computational Data Science (Computer Science & Mathematics)

• GPA: 3.5/4.0, Dean's Scholarship Recipient

Gayatri Vidya Parishad College of Engineering, Bachelor of Technology Jun 2015 – May 2019 India

• Major : Mechanical Engineering

• CGPA: 8.63/10.00, Full-ride scholarship

Contact this candidate