Data Scientist | Machine Learning Engineer | AI/ML Enthusiast

I'm a passionate Data Scientist with experience in building models, optimizing algorithms, and delivering impactful data solutions. Currently working at Amgen, I bring machine learning solutions to life, focused on real-world challenges and creating data-driven insights.

• 🎓 Pursuing my MSCS at Cal Poly Pomona with a focus on AI/ML
• 🌱 Currently expanding my expertise in JAX, CUDA libraries, and vector databases
• 🧑‍💻 Actively exploring open-source contributions to enhance and reflect my knowledge in ML
• 💼 Open to collaborating on AI/ML projects, especially those involving anomaly detection, NLP, or predictive modeling

• Programming: Python, PySpark, SQL
• Machine Learning & Data Science: TensorFlow, PyTorch, scikit-learn, JAX, RAPIDS cuML, OpenCV
• Cloud & Data Engineering: Azure Databricks, Apache Spark, AWS (EC2, S3, Lambda, RDS)
• GenAI & NLP: OpenAI API, Hugging Face, LangChain
• Data Visualization: Plotly, Power BI, Tableau, Qlik
• Development & Operations: GIT, Docker, CI/CD

• Anomaly Detection on Employee Devices: Built and deployed an anomaly detection model to monitor app crashes on workplace laptops using Databricks, PySpark, and Spark-NLP.
• Crop Suitability Modeling: Leveraged CNNs and satellite data to predict crop yield and suitability as part of my thesis in precision agriculture.
• Digital Assets Management Platform (DAMP) Image Analytics Pipeline: Led the end-to-end development of an image analytics pipeline, enhancing operational efficiency by 20% through FastAPI and advanced features like InceptionV3 classification, OCR, and geo-tagging for optimized inventory management.
• Data Management and Azure Deployment: Designed structured data storage solutions using Azure Blob Storage and integrated insights into Azure Data Lake, reducing data retrieval times by 40% and enabling better asset organization within DAMP.
• Consumer Experience Optimization with NLP: Developed an NLP pipeline to update users on laundry machine availability dynamically, reducing user query resolution time by 15% and enhancing the consumer experience.
• Predictive Modeling for Business Intelligence: Created predictive models for customer churn prediction, sales forecasting, and inventory optimization, achieving a 10% accuracy improvement by applying advanced statistical and ML techniques with RAPIDS cuML.
• Azure Cognitive Services Integration: Automated sentiment analysis and image recognition by integrating Azure Cognitive Services, reducing manual analysis time by 30% and boosting data accuracy by 20%.
• Interactive Qlik Dashboards: Built dashboards to visualize complex analytics, cutting decision-making time by 20%, driving a 5% increase in project success rates, and enhancing overall operational efficiency.

• Email: [email protected]
• LinkedIn: linkedin.com/subham-panda
• GitHub: github.com/SubhamCPP

"In a world of data, let's turn information into insight."

Thanks for stopping by! ✨ Feel free to reach out if you’d like to collaborate or chat about all things data!