Machine Learning Data Scientist

Vivek Bhavsar

Fremont, CA Phone: 341-***-**** Email: *****.*********@*****.***

LinkedIn: linkedin.com/in/vivekubuf GitHub: github.com/vbhavsar16 Summary

Data Scientist with a Ph.D. in Computation and Applied Mathematics and 5+ years of experience in Data Science, Data Analysis, and Software Development. Highly proficient in machine learning, numerical modeling, and cloud- based solutions, with expertise in Python, Scikit-learn, PyTorch, AWS, and Azure. Proven track record in developing predictive models, optimizing CI/CD pipelines, and delivering data-driven solutions that improve operational efficiency by up to 70%. Adept at working in fast-paced, cross-functional environments, with strong skills in managing ML pipelines, automating model deployment, and driving innovation in data-driven projects across industries. Committed to applying advanced analytical techniques to solve complex problems and improve decision-making processes. Technical Skills

BASH Script C/C++ CUDA Databricks Git Hadoop Kubernetes MATLAB NumPy Pandas Power BI ETL PyTorch Python Scikit-learn SQL Tableau TensorFlow Transformers Apache Spark AWS Azure GCP R PostgreSQL Lambda Kinesis S3 SageMaker Agile Azure CI/CD Machine Learning NLP SpaCy/NLTK Experience

Data Scientist May 2024 - Present

Coherent Corp. Fremont, CA

● Developed and deployed a wafer yield analysis and optimization application, leveraging Failure in Time (FIT) and Defective Parts Per Million (DPPM) metrics to improve production quality.

● Implemented and optimized a time-to-failure (TTF) predictive model, using multiple regression techniques with Scikit-learn, leading to more accurate power variation analysis over time.

● Increased wafer testing efficiency by 70% through automation and Python-based application development, replacing traditional methods and significantly accelerating testing processes. Machine Learning Engineer Feb 2024 - Present

Qwarke, Inc. St. Petersburg, FL

● Managed the AWS machine learning pipeline for a scholarly-focused social media platform, utilizing AWS Step Functions, Lambda, Kinesis, PySpark, S3, and SageMaker to streamline data processing and model deployment.

● Developed and implemented a recommendation algorithm combining content-based and collaborative filtering, improving video suggestions by aligning them with users' fields of expertise and interests, resulting in enhanced user engagement.

● Collaborated with cross-functional teams in an Agile, fast-paced MLOps environment, delivering scalable features and robust platforms to support continuous integration and deployment of machine learning models. Machine Learning Engineer May 2020 - Jan 2023

Saltriver Infosystems Gujarat, IN (Remote)

● Built and maintained CI/CD pipelines to streamline development and health score for the predictive model releases on Azure ML workspace.

● Trained the model with the large dataset of the test results, demographic, itinerary, and geographic data of the guests managed on Azure SQL.

● Automated model monitoring for predictive analytics, enhancing healthcare solutions for a hospitality client, resulting in 27% less employee emergency leave due to COVID-19. Projects

CropCatalyst - Reboot the Earth Challenge, United Nations and Salesforce github.com/vbhavsar16/CropCatalyst

● Collected and curated agrometeorological data to predict crop yield in tons per hectare, utilising Scikit-learn to train ensemble models with the linear regressor, SVM, Random Forest, XGBoost, LGBM, and CatBoost.

● Analysed yield gaps and developed optimization strategies with controllable parameters, providing actionable suggestions to farmers via a user-friendly app in native language.

● Developed a detailed roadmap for the app’s funding and sustainability, highlighting its long-term impact and viability.

● Collaborated with a team to deliver a comprehensive solution, demonstrating strong project management skills. Flood Semantic Segmentation Model using NVIDIA Toolkit

● Developed a Disaster Risk Monitoring (DRM) system using computer vision and ML techniques for image segmentation.

● Loaded satellite radar images and performed data augmentation using the NVIDIA Data Loading Library (DALI).

● Trained the model using U-Net with NVIDIA's Train Adapt Optimise (TAO) on NVIDIA GPU Cloud (NGC), then deployed it to the Triton Inference Server for transfer learning.

● Implemented near real-time analytics on edge devices at remote locations to achieve faster response times. Education

Data Analytics and Machine Learning Certificate of Completion Sep 2024 - Present ElevateMe Bootcamp

150+ hours of hands-on coursework, having completed a Data Analytics project and Machine Learning project using Tableau, Python, Machine Learning and SQL

University at Buffalo, SUNY Sep 2018 - Jan 2024

Doctor of Philosophy in Computation and Applied Mathematics University at Buffalo, SUNY Sep 2016 - Aug 2018

Master of Science in Mechanical Engineering

Gujarat Technological University June 2012 - Jul 2016 Bachelor of Engineering in Mechanical Engineering

Contact this candidate