Priyanshu Shekhar Sinha

Specialization Course on Natural Language Processing

2021

Deeplearing.ai and Coursera.

• Classification and Vector Spaces.
• Probabilistic Models.
• Convolutional Neural Networks.
• Sequence Models.
• Attention Models.

Post Graduation Program

2019

Great Lakes Institute of Management.

Major: Data Science Engineering

• Python
• Machine Learning
• Statistics
• Data Analysis
• Natural Language Processing
• Data Visualisation
• Deep Learning
• Feature Engineering

Specialization Course on Deep Learning

2018

Deeplearing.ai and Coursera.

• Improving Deep Neural Networks.
• Structuring Machine Learning Projects.
• Convolutional Neural Networks.
• Neural Networks and Deep Learning.
• Sequence Models.

Python Data Structures

2018

University of Michigan and Coursera.

Python

• Data Structures
• Object Oriented Programming

Bachelor of Technology

2014 - 2018

Sir M Visvesvaraya Institute of Technology

Major: Electrical & Electronics Engineering

• Electrical Machine Designing
• Engineering Mathematics
• Analog Electronic Circuits
• Electric Power Generation
• Logic Design
• Microcontrollers
• Power Electronics
• Control System
• Transformers and Induction Machines
• Field Theory
• Network Analysis

Senior Data Scientist

2023 May - Present

Acuity Knowledge Partners, Bangalore, IN

• Lead a team of 4 members in the design, development, and implementation of a highly flexible and user friendly Analysis Dashboard for a leading Investing Firm. This framework helps client to measure whether a Investment is successful in terms of revenue generation.
• Co-Lead a team of 7 members in developing and deployment of an sophisticated Data Pipeline for easy migration to any platform.
• Technology Stack: Python, Streamlit, PySpark, Docker, AirFlow

Associate - Data Science

2021 Dec - 2023 Apr

TheMathCompany, Bangalore, IN

• Lead a team of 5 members in the design, development, and implementation of a highly flexible and user friendly A/B Testing Framework for a leading Pharmaceutical client. This framework helps client to measure whether a promotion is successful in terms of revenue generation.
• Co-Lead a team of 7 members in developing and deployment of an Analysing and Flagging pipeline for anomaly detection in their sales by taking seasonality into picture. This framework helps client to identify customers and pursue the customers with some incentives and increase their revenue. This framework helps client to increase their revenue by $125K in first quarter.
• Build a API Neural Network based on LSTM for forecasting number of days when a person will return to make a purchase. This model helps client in balancing their supply and demand with a time estimation. Model performance: R1-Score- 0.9012.
• Technology Stack: Python, Deep Learning, Machine Learning, Docker, AirFlow, Co.dx

Data Analyst

2021 May - 2021 Dec

Recruitment Smart Technology Pvt. Ltd., Ahmedabad, IN

• Designed a scalable solution to generate periodically opportunity report, operational report and advanced analytics report on the product performance for different clients.
• Reduced time for new client integration from 10 days (manual) to 3 hours i.e. 97% improvement.
• Reduced time for generating operational report (5 days) and opportunity report (1 day) to 30 minutes i.e. 95% improvement.
• Reduced time for generating advanced analytics report from 7 days to 15 minutes i.e. 99% improvement.
• Automated the Power BI dashboard for regular data updation.
• Awarded as Star Performer of the Month.
• Technology Stack: Python, Data Analysis, Machine Learning, AWS S3, EC2, SES, Power BI, Docker

Data Scientist

2019 Aug - 2021 Apr

Sumyag Datascience Pvt. Ltd., Bangalore, IN

• Designed and implemented an end-to-end scalable solution for extracting information from unstructured documents for enterprise clients.
• Achieved cross-domain compatibility by designing a highly configurable data-driven pipeline for domains like commercial insurance, healthcare, automobile, real estate and telecommunication.
• Leveraged AWS auto-scaled infrastructure to build custom statistical machine learning models for generating embeddings to predict entities from a corpus of 60000+ documents, roughly summing up to a terabyte.
• Technology Sack: Python, NLP, Deep Learning, Machine learning