Ashyam Zubair
Data Scientist
Chennai,TN
Summary
Experienced Data Scientist with a strong background in data analysis and machine learning, skilled in software development and proficient in Python and SQL. Proven track record of delivering successful data-driven solutions in fast-paced environments, with strong communication skills and the ability to manage multiple priorities. Implemented solutions resulting in improved efficiency and decision-making for clients including the Chief Minister of Tamil Nadu and Dubai Police.
Overview
5
5
years of professional experience
5
5
years of post-secondary education
Work History
Data Scientist
Peninsular Research Operations
Chennai, TN
09.2021 - Current
- Developed and deployed stacked LSTM model to forecast commodity prices, improving decision-making for stakeholders in agriculture industry. One challenge in this project was obtaining high-quality and sufficient data for training the model. To address this, I implemented data cleaning and preprocessing techniques and worked closely with subject matter experts to identify relevant data sources.
- Created robust, scalable framework for extracting and organizing e-paper content for media monitoring. One major challenge in this project was the need to scrape data from websites at scale. To overcome this, I designed the framework to be modular and flexible, allowing it to adapt to changing data sources and structures. We also implemented a distributed approach using Azure services to split the scraping task into smaller components, which increased the speed and reliability of the process.
- Automated a complex ETL pipeline using Azure Functions and Event Hubs to collect and analyze hundreds of data points daily from various domains. One challenge in this project was ensuring the reliability and scalability of the pipeline as the volume of data increased. To optimize performance, we chose to use Rockset as the data warehouse solution because of its built-in listener for Event Hubs.
- Developed a task verification system using OpenCV to match images uploaded by users with a set of templates. The system achieved an accuracy of approximately 75% in image matching and also included OCR to read and match necessary text. One challenge in this project was finding the right method to match templates, which was overcome by implementing various image preprocessing techniques.
Data Scientist
Algonox Technologies
Hyderabad, TG
11.2017 - 09.2021
- Developed and deployed a classification model for the Dubai Police to improve the efficiency of the policing process. Utilizing historic crime data, the model was able to classify crimes with an overall accuracy of ~73%. This project had a significant impact on the effectiveness of the police force and helped to streamline their operations.
- Developed algorithms to extract various fields from unstructured PDF documents. One challenge in this project was the lack of consistent formatting and structure in the PDFs, which made it difficult to accurately extract the desired data. To overcome this, I implemented advanced pattern recognition techniques and worked closely with subject matter experts to identify key features and improve the accuracy of the algorithms.
- Automated the generation of accurately populated case summary templates from free text, reducing manual summarization time by 200%. One challenge in this project was the need to accurately parse and interpret the free text to extract the relevant information. To overcome this, I designed the system to use natural language processing techniques and worked with a team of linguists to fine-tune the performance of the system.
Education
Master of Science - Data Science
Liverpool John Moores University
Liverpool, England 09.2021 - 06.2022
Post Graduate Diploma - Data Science
International Institute of Information Technology
Bangalore, India 09.2019 - 09.2020
Bachelor of Engineering - Computer Science
Birla Institute of Technology And Science
Dubai, United Arab Emirates 08.2014 - 08.2017
Skills
Python
Data Analysis
Statistics
SQL
Data Visualization
Data Modeling
Data Mining
Natural Language Processing
Data Manipulation
Timeline
Data Scientist
Peninsular Research Operations
09.2021 - Current
Master of Science - Data Science
Liverpool John Moores University
09.2021 - 06.2022
Post Graduate Diploma - Data Science
International Institute of Information Technology
09.2019 - 09.2020
Data Scientist
Algonox Technologies
11.2017 - 09.2021
Bachelor of Engineering - Computer Science
Birla Institute of Technology And Science
08.2014 - 08.2017