Machine Learning Engineer

Mayank Sehrawat

Boston, MA ********.*@************.*** +1-857-***-**** LinkedIn Github

Education

Northeastern University, Masters in Applied Machine Intelligence Sep 2024 – Present Indraprastha Institute of Information and Technology, Delhi, Bachelors of Technology in Computer Science and Applied Mathematics Jul 2016 – Aug 2020

Experience

Data Science Consultant, DataisGood Delhi, India Feb 2024 - Aug 2024

• Designed forecasting analytic models using SAS, Scikit-learn, improving platform and boosting engagement by 60%. Led cross-functional collaboration, ensuring seamless deployment to achieve business goals.

• Built ETL pipelines using Python and SQL on Azure, enabling real-time insights. Leveraged Machine Learning algorithms to prepare dashboards in Tableau and Power BI, delivering data-driven recommendations. Computer Vision Engineer (R&D), Falcon Autotech Delhi, India Dec 2021 - Dec 2023

• Developed and optimized 6D pose estimation models using Python, C++, PyTorch, and YOLOv5 which uses CNN, integrating 3D cameras for real-time robotic pick-and-place tasks, achieving a 70% increase in efficiency.

• Engineered multi-modal fusion, combining robotic telemetry, OpenCV, TensorFlow, CNNs, and 3D vision, enhancing localization accuracy in dynamic environments. Designed and optimized synthetic data pipelines for data augmentation and preprocessing, improving detection robustness and adaptability.

• Initiated and deployed scalable MLOps pipelines on Azure, integrating REST APIs, SQL, cloud-based model deployment, and real-time inference. Collaborated with cross-functional teams to optimize the algorithm using image processing, object tracking, and deep learning, improving task execution and system efficiency. Data Analyst Intern, Getboarded Technologies Remote Jul 2021 - Nov 2021

• Conducted in-depth market trend analysis, risk assessment, and business intelligence modeling using Pandas, SQL, Scikit-learn, and statistical techniques, identifying growth and optimizing strategic planning.

• Leveraged Tableau and Power BI, enhancing data visualization and accelerating executive decision-making. Skills

Programming Language : Python, SAS, Java, C++, R

Machine Learning Frameworks: TensorFlow, PyTorch, Scikit-learn, Keras Data Processing and Visualization: Pandas, NumPy, Matplotlib, Seaborn, Tableau, MS Power BI, Scipy Databases: SQL, MySQL, NoSQL, Query Handling, MongoDB, Relational Database Soft Skills: Leadership, Analytical and Problem Solving, Teamwork, Collaboration Competencies: Cloud Computing : Azure, NLP, LLM, VBA, Model Deployment, Linux, API Integration, Opencv, 3D Cameras, Generative AI, Statistics, MS Excel, VS, Git Research & Projects

Fitness Trainer using ML

• Defined and processed an ML-based fitness trainer for the quantified self, analyzing consumer accelerometer and gyroscope data collected from wearable devices during squats, chest presses, deadlifts, Barbell lifts, etc.

• Engineered features using Fourier Transform and Principal Component Analysis (PCA) to extract meaningful insights from raw sensor data and for monitoring the growth of model development.

• Created a classification model using machine learning algorithms such as Neural Networks, Random Forest, k-nearest clustering, and SVM, achieving 95% accuracy in exercise identification and recognition. Song Recommender System

• Conceived a personalized results oriented song recommendations system using Python, leveraging AI collaborative filtering and content based filtering techniques to predict user preferences.

• Optimized accuracy through data preprocessing, feature engineering, resulting in 15% improvement.

• Evaluated model performance and visualized user behavior via constant testing to boost recommendations. Scalable NLP Pipeline for Sentiment Analysis and Topic Modeling

• Authored an NLP pipeline using Hugging Face for sentiment analysis and spaCy for preprocessing, analyzing over 1 million e-commerce reviews with 90% accuracy.

• Implemented topic modeling with LDA to extract key customer insights, driving product improvement strategies.

• Optimized large-scale text data processing using PySpark, reducing pipeline runtime by 30%.

Contact this candidate