• I have worked as a freelancer on various Cloud, software development, data science, and machine learning projects. My expertise as a Data Scientist encompasses building and testing machine learning models, conducting data analysis, developing prediction systems, and working in areas such as computer vision, speech recognition, and speech generation. I also have experience in constructing data analysis APIs.• I have successfully completed over 95 large-scale projects on Upwork.• I am proud to hold the 'Top Rated Plus' badge, the highest accolade on the platform, with a 100% job success score

• Contributed to the development of PythonSDK.• Orchestrated the development of a web-application with backend and DevOps for creating machine learning training jobs, training data selection, and model saving.• Acted as a Python Developer crafting Extract, Transform, Load (ETL) scripts.• Managed real-time requests with optimized logic for timely execution.• Engineered advanced Flask API to manage AWS marketplace billing.• Pioneered machine learning models for face, voice, and fingerprint verification.• Executed hybrid Siamese Network implementation.• Implemented 1-way Fully Homomorphic Encryption.• Established Cloud Biometric Authentication via API and oAuth2.0.

• Designing algorithms and analysing data for psychometric tests.• Worked on developing a web-application including backend and DevOps part for creating machine learning training jobs, selecting training data, saving models and dashboard for selecting models for inference.• Created Fask APIs for inference endpoints.• Technologies Used: Sagemaker, Fask Framework, SQL, Python, AWS, DevOps, Nodejs, Redis

•Collaborated with a team to analyze AWS consumption datasets, focusing on resource utilization, costs, and metadata.•Conducted comprehensive research to identify suitable machine learning algorithms tailored to the dataset's characteristics.•Developed and fine-tuned ML models to discern patterns in resource tagging.•Successfully implemented an auto-tagging system that categorizes AWS resources based on historical tagging patterns, optimizing resource management and cost-efficiency.

• The goal of the research was to improve the probabilistic descriptions of hydrologic modellingerrors and attain better constrained parameter estimates . My task was to analyze data-drivenmethod for reliable estimation of predictive uncertainty and to create a Web based Applicationwhich generates uncertainty intervals using kNN resampling for a given time series of predictionsResearch paper- https://www.hydrol-earth-syst-sci.net/21/4021/2017/Tools/Skills : Javascript , Machine Learning ,Node.js