My Professional Journey

Sr Data Scientist  at DXC Technology                                2019–Continue

• Achieved 94% accuracy and 91% AUC in identifying fraud transactions by using the Ensemble approach on the cloud. 

• Helping clients for their digital transformation, migrating on-premise data solution infrastructure to scalable cloud resources.

• Developing AI solutions in Azure & Google cloud platform (GCP) using programming, cloud resources, open-source tools.

• Built batch and real-time stream data ingestion, developed and integrated ML with data platform using GCP & open source.

Research Assistant  at Cradle to Career Policy Institute, UNM        2017–2019

• Build data preparation framework to extract census data, resulting in a 70% decrease in time needed by using Python and R.

• Perform statistical learning method and inferential statistics to make prediction & implications of developing strategy.

• Develop interactive data analysis tools by using Tableau for New Mexico Office of African American Affairs which is used by policymakers for planning programs and allocating resources.

• Design Microsoft SQL server database and use SQL queries for data extraction and manipulation.

Network Analyst at Edotco Bangladesh Limited           2015–2016

• Developed machine learning method, managed 50 field service engineers and reduced 27% potential network incident.

• Executed clustering, tree-based method, and statistical learning to manage network more proactively by using Python and R.

• Derived actionable insights that can help management in decision-making & goal-setting by using analytics and visualization.

• Steered training all over the country for the competent development of engineers which was effective to achieve the target.

• Awarded Hero’s enabling connectivity for increasing network availability by using predictive model during the period 2015.

Service Engineer at L M Ericsson Bangladesh                2010-2015

• Performed statistical analysis of raw server data to locate the most vulnerable area of 2500 transmission nodes.

• Implemented supervised and unsupervised machine learning method to have a grip on the wireless network with Python.

• Used inferential statistics to find hidden insight from large volume server data from the wireless network by using R.

• Coordinated wireless project with 25 vendors, customers, and other stakeholders to achieve within timeline and budget.