Dell Technologies

• Working with cutting-edge technologies to deliver innovative solutions for the Infrastructure Solutions Group.

Micron Technology

• Developed firmware modules and improved code coverage by more than 20% with BullseyeCoverage.
• Migrated functionalities from the deprecated PIT framework to Python. Added support for 25+ features exercising commands on simulated and physical hardware over the NVMe bus.
• Implemented 5+ OEM-specific features as a part of the Open Compute Project (OCP) specifications, contributing to the development of industry standards.
• Refactored vendor-specific source code (5000+ LOC) and incorporated better coding standards to improve the codebase's maintainability.
• Devised unit tests in C++ using Google's Testing and Mocking Framework to improve code quality, running automatically during feature commits and reducing execution time by ~10%.
• Achieved 100% coverage and validation of NVMe Host and Controller Telemetry Log Pages, ensuring the functionality and reliability of the module.
• Tools: Python, C++, Enterprise Storage, Shell Scripting, Linux, Git, CI/CD, NVM Express (NVMe), Jenkins, Bitbucket, JIRA, Confluence.

Texas A&M University

• Conducted an in-depth analysis of patent litigation in the United States, using data analysis techniques to identify trends and patterns.
• Formulated a framework for automatically syncing data from Contactually CRM to Google Contacts using Google People API, saving over 3 hours of manual effort in each iteration and improving data accuracy.
• Demonstrated strong analytical skills and attention to detail in working with data, ensuring accuracy and reliability. Collaborated effectively with team members and stakeholders to develop and execute projects, contributing to achieving organizational goals.
• Tools: Python, Data Science, Machine Learning, Numpy, Pandas, Scikit-learn, NLTK, Gensim, Scipy, Jupyter Notebook, Matplotlib, Seaborn, Excel, GCP, Colab.

Texas A&M University

• Worked in the Perception, Sensing, and Instrumentation (PSI) Lab with Dr. Ricardo Gutierrez-Osuna, contributing to research efforts in bioinformatics.
• Developed a diagnostic platform for detecting early-stage pancreatic ductal adenocarcinoma (PDAC) in higher-risk patient cohorts, using data analysis techniques to identify potential biomarkers.
• Focused on identifying an assay panel for early detection of exocrine pancreatic cancer, particularly in stages I & II when surgery is still an effective therapy.
• Achieved a peak accuracy of 75% with the developed framework, demonstrating the effectiveness of the diagnostic platform.
• Collaborated effectively with research team members to develop and execute the project, contributing to achieving the overarching research goals.
• Tools: Python, Data Science, Machine Learning, Deep Learning, BioinformaticsFlask, Scikit-learn, Pandas, Plotly, Matplotlib, Colab, Kaggle.

Western Digital

• Led a 4-member team to verify storage protocols for next-gen client-class Solid State Drives, ensuring product quality and performance for Western Digital Corporation.
• Actively participated in defining, developing, and analyzing new firmware features, contributing to functional and protocol development, execution, and validation of WD NVMe SN730, WD NVMe SN530, and WD Blue.
• Conducted an in-depth analysis of firmware bugs caught during validation and studied the feasibility of test cases that could replicate such scenarios during Firmware verification, leading to better and more efficient firmware development.
• Created an integration pipeline using Jenkins and Gerrit Code Review, which accomplished a ~10% improvement in code deployment time and facilitated smoother code review processes.
• Implemented Hyperparameter Tuning with an internal test prioritization tool (OctoSpider), resulting in a 15% improvement in performance for the client-class Solid State Drives.
• Created a visual framework extracting features from Google's Blockly to interact with and validate NVMe Devices.
• Verified customer escalated issues and served as the sole point of contact for Oakgate and DriveMaster, addressing issues and providing timely resolution.
• Published research papers in the WDC IDTC on topics such as A Pub-Sub protocol-based validation framework and Zoned Storage, contributing to industry-wide research and development in storage technology.
• Tools: C, C++, Python, Client Storage, Storage Security, NVM Express (NVMe), Machine Learning, Flask, Jenkins, Gerrit, DriveMaster, Oakgate, MQTT, Scikit-learn, NLTK, Gensim, CI/CD, Shell Scripting, Linux, Git, Blockchain.

Codementor

• Coordinated with clients worldwide on freelance projects, long-term mentoring, and consulting related to Software Development, Data Science, and Machine Learning.
• Delivered successful solutions to 60+ clients worldwide, which resulted in a 5/5 aggregate rating.
• Maintained effective communication with clients throughout the project lifecycle to ensure timely delivery and quality output.
• Enhanced client satisfaction by addressing feedback and incorporating it into the project requirements.
• Profile - https://www.codementor.io/@dushyantbgs.
• Tools: Python, JavaScript, Flask, Django, Scikit-learn, Tensorflow, Keras, NLTK, Spacy, Gensim, Selenium, Spacy, MongoDB, AWS, MQTT, Raspberry Pi.

Delhi Technological University

• Collaborated with Dr. Akshi Kumar in conducting research on Deep Learning and Language Technologies.
• Utilized Neural Word Embeddings and Feature Extraction techniques for Feature Engineering.
• Developed Deep Learning models using Keras and TensorFlow frameworks for Genre Classification.
• Achieved a peak precision and f-score of 79.6% and 84.09%, respectively, surpassing the baseline models.
• Conducted research in Software Defect Prediction with Genetic Algorithms under the guidance of Dr. Ruchika Malhotra.
• Applied Differential Evolution and Simulated Annealing as optimizers on various datasets from open-source JAVA systems to explore the tuning space.
• Analyzed the results obtained from the optimization process to identify the best algorithm and parameter settings for software defect prediction.
• Acquired experience conducting research, analyzing experimental results, and presenting findings at conferences.
• Tools: Python, Tensorflow, Keras, Scikit-learn, NLTK, Gensim, Pandas, Numpy, Matplotlib, Colab.

Accelerity

• Designed and implemented an end-to-end architecture for IoT data ingestion, storage, and analysis.
• Developed RESTful APIs using Flask and Django to enable data retrieval and management. Integrated MQTT protocol to ensure seamless data flow from sensors to the platform.
• Utilized AWS Lambda to reduce infrastructure costs and optimize the performance of the IoT platform. Configured and optimized AWS S3 and DynamoDB to efficiently store and manage large volumes of IoT data.
• Implemented real-time data streaming using AWS Kinesis and explored its potential for machine learning and predictive analytics.
• Collaborated with front-end developers to ensure seamless integration of the backend with the web application.
• Tools: Python, Flask, Django, MQTT, AWS Lambda, AWS S3, AWS DynamoDB, AWS Kinesis, JavaScript, Scikit-learn.

SanDisk

• Worked with the Client Storage Solutions team at the SanDisk India Device Design Center.
• Developed test plans, test cases and executed test scenarios for validating OPAL and other security-related features of Solid-State Drives.
• Developed a tool using Python that automated the test process and reduced the execution time by 10%-15%.
• Tools: Python, DriveMaster, TCG, OPAL.

Indiabulls Housing Finance

• Spearheaded the development of the Leave Management Platform, from requirements gathering to deployment, ensuring seamless tracking of employee leave requests, approvals, and management, improving operational efficiency by 25%.
• Worked with HTML, CSS, JavaScript, and jQuery to design and develop a user-friendly interface for the Leave Management Platform, enhancing the user experience. Employed Bootstrap to develop a responsive design for the web platform, optimizing the platform for various screen sizes and devices.
• Utilized PHP to create server-side scripts for processing data and seamless communication with the MySQL database.
• Leveraged Python and Scikit-learn to develop a machine learning model that predicts stock price movements using daily news, providing valuable insights for stock traders.
• Tools: Python, Scikit-learn, HTML, CSS, Jquery, Bootstrap, PHP, JavaScript.

Languages

Python

C/C++

JavaScript

HTML

CSS

PHP

SQL

Shell Scripting

Frameworks, Tools, and Libraries

Django

Flask

Bootstrap

Tensorflow

Keras

PyTorch

NLTK

Gensim

Scikit-learn

Pandas

Numpy

Matplotlib

Plotly

MySQL

MongoDB

Jenkins

Git

MQTT

Cloud Computing and Orchestration

AWS

Heroku

Docker

Kubernetes