Neeraj
New Delhi, Delhi, India
Ó +91-9871520187
[email protected] Linkedin Github Portfolio
SUMMARY
Machine Learning & MLOps Practitioner with experience in building scalable AI solutions. Skilled in deploying
models using AWS, Docker, and CI/CD pipelines. Passionate about automating ML workflows and optimizing
model performance for real-world applications. Experienced in ML model orchestration, experiment tracking, and
cloud computing.
SKILLS
Languages: Python, SQL (PostgreSQL)
Data Science: A/B Testing, Hypothesis Testing, Data Science Pipeline(Cleaning, Wrangling, Visualization,
Modeling, Interpretation), EDA, Statistics
Machine Learning: Python(e.g. scikit-learn, numpy, pandas, matplotlib), Machine learning Pipeline(Data
Ingestion, Transformation, Validation, Model Training and Evaluation), ETL, Feature Engineering
MLOps: Apache Airflow, MLflow, CI/CD (GitHub Actions), Docker, FastAPI, Flask, Grafana
Cloud Technologies: AWS (EC2, S3, ECR, Beanstalk, SageMaker), Azure (Web App, Container Registry)
Tools/Technologies: Git, GitHub, Jupyter Notebook, Zenml, VS Code, APIs, Streamlit, DVC
PROJECTS
Malicious URL Detection System OCT 2024 - Dec 2024
• Developed an end-to-end MLOps solution for malicious URL detection, achieving 97.6% F1-score.
• Tech Stack: Python, MongoDB, MLflow, Airflow, FastAPI, Docker, GitHub Actions, AWS (S3, ECR, EC2)
• Built a FastAPI backend for model training and batch predictions, reducing inference time by 35%.
• Implemented an automated model training pipeline and prediction pipeline using Apache Airflow, scheduled
weekly to ensure regular model retraining.
• Tracked experiments using MLflow, hosted on AWS EC2, with artifacts stored on AWS S3.
• Performed Data validation by ensuring schema correctness, detecting data drift, & generating drift reports.
• Deployed the application on AWS EC2 using Docker containerization and the Docker image stored in ECR.
• Implemented CI/CD pipeline using GitHub Actions for automated deployment of application to AWS EC2,
reducing deployment time by more than 30%.
Machine Predictive Maintenance Classification JAN 2025 - FEB 2025
• Developed an end-to-end MLOps solution to predict machine failures, achieving an F1-score of 96.4%.
• Tech Stack: Python, Flask, Docker, GitHub Actions, AWS (Beanstalk, EC2, CodePipeline), Azure, Xgboost
• Evaluated ML models to predict machine failure types, selecting the best model based on F1-score.
• Optimized model performance through hyperparameter tuning using GridSearchCV.
• Deployed the application on AWS EC2, AWS Elastic Beanstalk, and Azure web app, ensuring scalability,
reliability, and 99.9% uptime in production.
Customer Satisfaction Prediction SEP 2024 - SEP 2024
• Built an end-to-end ml pipeline for continuously deploying the machine learning model and prediction,
alongside a data application that utilizes the latest deployed model for the business to consume.
• Tech Stack: Python, ML Pipeline Orchestration (ZenML), Model Deployment(MLflow), Streamlit
• Built a training pipeline using ZenML and MLflow for experiment tracking and model deployment.
EDUCATION
Netaji Subhash University of Technology, New Delhi, DELHI 2020 – 2022
Completed Coursework: Calculus, Intro to Computer Programming, Algebra, Probability Theory.
ACHIEVEMENTS
• Created and shared technical content on AI, ML, and MLOps, building an engaged community of 7K+ tech
professionals and practitioners in 8 months, leading to Recognition from established professionals in the field.
