Los Angeles, California, United States
AI Developer | Data Scientist I am an AI developer who works to develop ingenious solutions.
– Developed a T5+ based code summarization app to convert code snippets into summaries, streamlining the code reviews. – Developed an GPT-based ticket management system for automatic ticket categorization and solution generation. – Engineered robust data pipelines in Cribl, efficiently managing daily data ingestion of 15TB. – Evaluated infrastructure health leading to a 30% reduction in infrastructure-related issues. – Modernized legacy infrastructure by expanding cloud utilization, establishing data pipelines to Amazon S3. – Developed data enrichment mechanisms using Splunk to enhance operational visibility of logs and metrics. – Orchestrated the development of log ingestion pipelines using Splunk, enabling observability use-cases across platforms.
– Honed fervor for speaking-emoting-expressing. – Pursued the path of 'Dynamic Leadership' which pertains to handling people, resolving conflicts, guiding others into the correct direction, self-reflecting and incorporating candor in daily life.
– Became Vice President Membership in June 2020 and functioned as the HR manager of the club by handling ongoing routines pertaining to new member inductions, inter-club bonding and member satisfaction.
– Developed a Burnout Vulnerability Prediction feature to estimate an individual's potential for burnout using static and dynamically recorded data. – Utilized XGBoost as the model to analyze various factors like workload, stress levels, and behavioral indicators to predict burnout vulnerability. – Trained the model on both static and dynamically recorded data to capture a more comprehensive view of an individual's burnout risk – Deployed the feature using Flask, making it user-friendly and accessible to individuals and organizations seeking to manage burnout risk.
– Developed a similarity-based mentor-student matching system based on a dataset of 1,000 profiles, enhancing mentorship effectiveness by 45%. Achieved a 90% match accuracy. – Incorporated user inclinations to specific attributes by assigning decision weights, enhancing personalization and increasing user engagement by 30%. – Implemented various similarity measures (cosine similarity, Euclidean distance) and conducted A/B testing to validate the matching system’s effectiveness. – Designed and deployed a user interface using Flask, ensuring accessibility with 99% uptime.
– Preprocessed raw data related to customer behaviour and transaction records, ensuring high-quality inputs for analysis. – Developed scalable machine learning pipelines using Sklearn, Pandas, Numpy, and Scipy, streamlining ML workflows. – Automated data cleaning activities, reducing manual effort by 95% and improving data consistency. – Optimized model performance by conducting hyperparameter tuning using grid search and Bayesian optimization. – Documented and communicated results and insights to stakeholders to increase business efficiency.