Akash Yanampally

Designed and automated end-to-end ML and Generative AI pipelines using MLflow, Airflow, Kubeflow, and Vertex AI across multi-cloud environments; developed and fine-tuned LLM and NLP models to power intelligent chatbots, automated regulatory document analysis, and context-aware advisory assistants; built high-throughput data streaming and analytics systems; deployed forecasting and predictive models; implemented vector-based retrieval and RAG with FAISS, Pinecone, and Weaviate; enhanced explainability and governance with SHAP, LIME, Fairlearn, and LangSmith; delivered BI dashboards translating model outputs into executive insights; containerized AI microservices with Docker/Kubernetes and automated CI/CD; designed secure data governance with RDF/graph technologies.

Developed and optimized end-to-end ML pipelines using TensorFlow, PyTorch, and Scikit-learn for predictive analytics in healthcare-like contexts and supply chain optimization; leveraged distributed data systems (Spark/Hadoop/Databricks) for multimodal data; implemented real-time data ingestion with Kafka and Airflow; deployed scalable models on Azure ML and GCP Vertex AI with MLOps practices; enhanced interpretability with SHAP/LIME; built dashboards for stakeholder insights; containerized services with Docker/Kubernetes and CI/CD pipelines.

Designed large-scale ML models for demand forecasting and optimization; automated data ingestion/transformation with Airflow, Spark, and Pandas; built real-time streaming with Kafka/Flink; developed analytics pipelines for risk and operations; deployed models via containerized microservices with Docker/Kubernetes/Helm; implemented observability with Prometheus, Grafana, and ELK; delivered business-ready dashboards and REST/GraphQL APIs for decision support.