Srushti Manjunath Email & Phone Number

As a dedicated Data Scientist with a strong background in optimization and industrial engineering, I have honed my skills across various sectors, including finance, technology, and e-commerce. Recently, I have been focusing on generative AI and data-centric graph learning, exploring methods to improve data quality, develop cutting-edge AI models, and enhance predictive analytics.During my master's at the University of Illinois, I strengthened my expertise in ML, DL, RL, network analysis, and GNN, applying these concepts to real-world challenges. My professional journey includes significant roles at Wells Fargo, where I spearheaded generative AI projects using TensorFlow and PyTorch, and Genpact, where I developed a Python-based chatbot and built efficient data pipelines using AWS. At Amazon India, I improved demand forecast accuracy and led process redesign initiatives.I am certified in AWS Cloud, TensorFlow, and NLP, and I continue my research on advanced graph-based deep learning for predictive analytics, focusing on enhancing demand prediction accuracy using GNNs. My current projects include leveraging spatio-temporal features and causal inference to optimize forecasting accuracy.Through these experiences, I bring a blend of academic rigor and practical industry application to the table, ready to contribute to diverse teams and projects. Let’s connect and explore how we can leverage data to create impactful solutions!Skills:• Programming and Scripting: Python, R, SQL, MATLAB, C++, Scala, React JS, RESTful APIs, Flask• Libraries and Frameworks: PySpark, PyTorch, TensorFlow, Scikit-learn, NumPy, Matplotlib, Keras, SpaCy, NetworkX• Cloud: AWS, Databricks, Glue, Lambda, S3, EC2, Redshift, Athena, Bedrock, Kendra, Sagemaker• Machine Learning: Supervised and Unsupervised learning, neural networks, CNN, RNN, Transformers• Big Data and Database Management: Apache (Hadoop, Spark, Kafka), MapReduce, MongoDB, MySQL• LLM and Other Platforms: GPT, BERT, BART, PaLM, LLaMA, Claude, Docker, Tableau, Power BI, Azure DevOps• Graph Learning and Knowledge Graphs: GNN frameworks and architectures, graph embeddings, graph generation, Neo4j, Amazon Neptune, RDF, SPARQL, Predictive Analytics with Graphs, Data Quality Improvement

• 

  Data Scientist

  Wells Fargo

  May 2023 - Present - United States

  • Spearheaded the development and implementation of generative AI models, including GPT and LLaMA, with TensorFlow to generate human like text, enhancing natural language generation capabilities.• Leveraged TensorFlow and PyTorch for developing custom neural networks with 95% accuracy for fraud detection, using data augmentation and transfer learning to reduce training time and improve resource efficiency by 30%.• Employed foundational models like Claude and LLaMA for text generation and reasoning, creating custom AI applications for personalized financial advice using Agents and Knowledge base from AWS Bedrock.• Enhanced customer service automation by fine-tuning models with LoRA, PEFT, and RLHF for accurate generative AI.• Contributed to the development of a GenAI tool by leveraging LangChain for chaining language models and RAG for advanced information retrieval and processing.• Collaborated with cross-functional teams and stakeholders to develop statistical and NLP solutions, building and integrating supervised learning models with scikit-learn, SparkML, and spaCy in Agile pods to support credit scoring, marketing, and customer segmentation projects.• Integrated models with web applications to enable real-time interaction and usability across business workflows.• Utilized Docker to containerize the models, ensuring portability and scalability in deployment across diverse environments.