Abdul Sohail Ahmed

Proficient in data collection, preprocessing, and analysis, with a strong ability to extract actionable insights from complex datasets. Skilled in utilizing advanced statistical and machine learning techniques to drive informed business decisions. Adept at communicating findings to both technical and non-technical stakeholders, fostering collaboration, and driving data-driven strategies. Eager to apply analytical skills, domain knowledge, and passion for problem-solving to contribute effectively to a dynamic data science team.

1. Wine Quality Classification (Languages & Tools: Python, Tableau)

• Spearheaded the development of a data-driven model for wine quality assessment, merging physicochemical and sensory data to ensure comprehensive analysis.
• Employed advanced data visualization to uncover patterns, enhancing the accuracy of the quality assessment model and providing actionable insights.
• Collaborated with wine manufacturers, implementing the standardized quality evaluation model to enhance production processes and consumer satisfaction.
• Facilitated data-driven decision-making by effectively communicating complex findings to both technical and non-technical stakeholders.

2. Analyzing H1B Visa Applications (Languages & Tools: Python, MongoDB, MySQL, PostgreSQL, Tableau, Azure CosmosDB)

• Orchestrated project focusing on H1B visa application data analysis using a specialized Database Management System, aiming to enhance transparency and accountability within the program.
• Engineered a comprehensive schema to decipher key insights like top job roles, sponsoring employers, wages, and location influence on visa outcomes, leveraging SQL Queries, Aggregation pipelines, and data warehouse principles.
• Pioneered an interactive Tableau dashboard to visually represent data trends, empowering informed decision-making and fostering engagement with the H1B visa program.

3. Lyrics Generator using LSTM Concept (Language: Python (Keras, Pandas, Numpy and Tensorflow)

• Led a project centered around Recurrent Neural Networks (RNN) in tandem with Long Short Term Memory (LSTM) architecture, innovatively designed to generate diverse and user-relevant song lyrics.
• Developed an interactive system that takes user-provided starting lyrics and employs advanced RNN-LSTM techniques to craft a range of lyric variations, enhancing user engagement and satisfaction.
• Successfully harnessed the power of machine learning and natural language processing to create a dynamic platform capable of generating personalized song lyrics, showcasing a fusion of creativity and technical expertise.