Keshav Sharma is an experienced Machine Learning Engineer with expertise in Computer Vision and Data Engineering. His background includes developing models for prostate cancer assessment and image deblurring using deep learning techniques. He has practical experience utilizing cloud platforms like AWS and GCP, and frameworks such as TensorFlow and PyTorch.
Experience
Machine Learning Engineer
Quantiphi
Computer Vision: Worked on WSI images related to healthcare and loaded a large dataset with images and masks for PANDA- Prostate Cancer Grade Assessment. Trained models using SEResNeXt 50, EfficientNet, and DenseNet models, measuring performance with QWK score, Accuracy, and loss curves. Worked on image deblurring, utilizing deep learning (Autoencoders, GAN’s) and non-deep learning approaches (Wiener Filter, Richardson-Lucy) optimized with TensorRT, ONNX runtime, and Pruning. Data Engineering: Worked with RDL CIS team for maintenance and modification of Data Transfer API, Workbench API, and Data Ingestion. Used GCP for testing cloud functions and implemented Pytest and Pydantic modules in Python with Flask to validate incoming requests.
Technology Intern
Snowflake
Gained knowledge of AWS and DBT concepts. Started with Threat Detection and wrote Detection Improvement Requests. Collaborated with the Platform automation team to implement GZIP compression in AWS Lambda response for API calls. Developed proficiency in MySQL and Python programming languages, focusing on Security, AWS, and Snowflake platforms.
Education
MIT AOE, Pune
B.Tech
Computer Engineering
Boys High School, Jammu
XII
Licenses & Certifications
Google Cloud Certified - Associate Cloud Engineer
AWS Certified Solutions Architect – Associate
Data Science Specialization
IBM
Web development Bootcamp
UDEMY