
PERSONAL STATEMENT A methodical data analytics and scientiest seeking an opportunity with an organization to leverage my skills. Looking to use existing knowledge and my experience of Data Science, ETL, Reporting, Business analytics in a professional setting.
Enterprise Independent Testing
Creating workflows using trifacta as per the business requirement. Fetching the raw data from external vendors
via emails and writing codes to fetch from various external environments such as SQL, Teradata, SAS EG etc.
Generating population flows, and sample flows using various sampling methods such as general random and
statistical along with the test result detail flows and test sum flows using public connections, and private
connections in trifacta for the bank. Performing data
validation, debugging issues and testing the samples for the compliance tests and providing them to our testing partners for further testing.
Creating artifacts and connecting with the LOB (Line of Business) for further discussions on new Business
requirements for sample creation and testing.
IDFC FIRST Bank
Monitoring Daily, Weekly and Monthly SAS MIS reports. Debugging the issues in case of failures, modifying
the reports. Responsible for providing the insights on the basis of financial analysis and generating reports to
represent it in higher level meetings for the business strategic planning.
Risk analysis to further calculate the corrective and preventive measures for the better business planning.
Client: Merck-Serono
Strong knowledge of global clinical trial practices, procedures, development process, and clinical trial data flow, especially in Oncology.
Wrote SAS and in trifacta data programs for management, merging, and analysis for every aspect of clinical reporting including datasets and listings. Used SAS/SAS SQL functionality used in clinical trial reporting including the Macro, BASE SAS.
Designing, programming and testing SAS programs as per requirement in accordance with programming standards
and validation procedures
Client: Core SAS Projects
Have worked as lead programmer on clinical studies of all phases in therapeutic areas like COVID, Respiratory disease etc. Handled different SAS activities during set-up, conduct and close-out stages of a study in RAVE/Inform database.
Drafted data transfer guidelines which includes all the information regarding datasets to be send to BIOS team, transfer mode, transfer frequency and quality and testing strategies.
Have set-up data transfer program which comprises of merging coding datasets from coding library with raw datasets and other study specific requirements to generate transfer datasets which are then send to biostaticians.
Have worked on data import program to reconcile the vendor data received is per Non-CRF guidelines. Analysed clinical trials data and generated DM Listings, MDR listings, Vendor recon listings (LAB,ECG,PK) , SAE recon listing according to Specifications.
Reviewed SAS specifications and non-CRF guidelines per sponsor requirements
Client: Moody’s Investor Service (Aug’17 – May’18)
Interacting extensively with end users on requirement gathering, analysis and documentation. Involved with key
departments to analyze areas and discuss the primary model requirements for the project.
Documented methodology, data reports and model results and communicated with the Project Team / Manager
to share the knowledge. Imported Data from relational database into SAS files per detailed specifications.
Carried out data extraction and data manipulation using PROC SQL, PROC SORT, PROC REPORT to create
preferred customer list as per business requirements Followed Software Life Cycle Development Models (SDLC) like Waterfall Model and Agile Model.
Client: Standard Life Investment (Jun’18 – Oct’19)
Analyzing and implementing the code as per client requirements. Implementing SQL pass through facility, data merging, concatenating datasets, sorting, macros, etc.
Communicating with clients to determine specific requirements and expectations; managing client
expectations as an indicator of quality. Unit Testing and debugging the errors.
Preparation of test plans and unit testing. Creating and managing the estimates, project plan, project schedule to
ensure that targets were reached on time.
70%
67.17%
86.18%
Spot Award: For delivery and effectively meeting the timeline (09/2018-09/2018) – Atos/Syntel
Impact Award: For valuable contribution towards the programming of custom listings and also meeting client expectations. (09/2020 – 09/2020) - Iqvia
Impact Award: Recognition from India MDR stakeholders for efforts and contribution in making MDR SAS Smart Listings initiative a success. (11/2020 – 11/2020) - Iqvia
Award and Recognition: For great work in achieving a 100% success rate in September, March and April for both delivering samples/results on time and accurately (09/2021, 03/2022 and 04/2022) - Bank of America
Indoor Scene Recognition- Used Tensor Flow, Keras - R Language
The task is to predict the type of room/scene present in the image. Indoor scene recognition is a challenging problem since some indoor scenes can be well defined by global spatial and structural properties, while others are better characterized by the objects included in the space. The dataset is a subset of a larger dataset for indoor scene recognition. The images are divided into train, validation, and test folders, each containing the folders related to the type of the room in the image (i.e. the categories of the target variable): bathroom, bedroom, children_room, closet, corridor, dining_room, garage, kitchen, living_room, stairs. The number of images available for each scene is variable and it ranges from 52 to 367 in the training set, with validation and test sets being roughly half the size.
Built a predictive model to predict the type of indoor scene from the image data.
Deployed Deep Neural Network (DNN ), Deep Neural Network with regulariser, dropout, data augmentation, Convolutional Neural Network and Convolutional Neural Network with data augmentation.
Compared appropriately the deep learning systems considered, evaluating and discussing their relative merits. Commented on their training and predictive performance, and select the best model a predicting the type of indoor scene from the data.
Used the test data to evaluate the predictive performance of the best model. Commented on the ability of the model at recognizing the different scenes.
Bringing Data to Life – Data Analysis of Housing Data - R language
Performed exploratory data analysis on the dataset.
Checked for missing values, errors and corruption.
Factor variable creation so that it yields better results in statistical modelling.
Plotted graphs to understand the trends, relationships between two variables.
Performed normalization on the data i.e. Zscore normalization so that it normalizes the outlier in the dataset in such a way that it's no longer a massive outlier.
Performed linear regression between the variables sqt and the prices to understand the pattern.
Detect Credit Card Fraud with Machine Learning in R
I hereby declare that the information that I have furnished is authentic and true to the best of my knowledge