Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI/ML Engineer and Data Scientist with 4+ years of experience building deep learning and LLM-driven systems across AWS and Databricks. I design scalable RAG pipelines, fine-tune sequence models, and develop end-to-end MLOps workflows for real-world healthcare and commercial use cases. I specialize in transforming large unstructured datasets into measurable outcomes through secure, HIPAA-compliant architectures, robust monitoring, and deployment automation. I enjoy bridging statistical modeling, deep learning, and production engineering to ship reliable ML products.…AI/ML Engineer and Data Scientist with 4+ years of experience building deep learning and LLM-driven systems across AWS and Databricks. I design scalable RAG pipelines, fine-tune sequence models, and develop end-to-end MLOps workflows for real-world healthcare and commercial use cases. I specialize in transforming large unstructured datasets into measurable outcomes through secure, HIPAA-compliant architectures, robust monitoring, and deployment automation. I enjoy bridging statistical modeling, deep learning, and production engineering to ship reliable ML products.

RAMU DEVARAPALLI

AI Engineer, Data Scientist, Software Engineer, +3





AI/ML Engineer and Data Scientist with 4+ years of experience building deep learning and LLM-driven systems across AWS and Databricks. I design scalable RAG pipelines, fine-tune sequence models, and develop end-to-end MLOps workflows for real-world healthcare and commercial use cases. I specialize in transforming large unstructured datasets into measurable outcomes through secure, HIPAA-compliant architectures, robust monitoring, and deployment automation. I enjoy bridging statistical modeling, deep learning, and production engineering to ship reliable ML products.…AI/ML Engineer and Data Scientist with 4+ years of experience building deep learning and LLM-driven systems across AWS and Databricks. I design scalable RAG pipelines, fine-tune sequence models, and develop end-to-end MLOps workflows for real-world healthcare and commercial use cases. I specialize in transforming large unstructured datasets into measurable outcomes through secure, HIPAA-compliant architectures, robust monitoring, and deployment automation. I enjoy bridging statistical modeling, deep learning, and production engineering to ship reliable ML products.

Available to hire

AI/ML Engineer and Data Scientist with 4+ years of experience building deep learning and LLM-driven systems across AWS and Databricks. I design scalable RAG pipelines, fine-tune sequence models, and develop end-to-end MLOps workflows for real-world healthcare and commercial use cases.

I specialize in transforming large unstructured datasets into measurable outcomes through secure, HIPAA-compliant architectures, robust monitoring, and deployment automation. I enjoy bridging statistical modeling, deep learning, and production engineering to ship reliable ML products.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Language

Aragonese

Intermediate

Bashkir

Intermediate

Work Experience

AI/ML Engineer at Humana

January 1, 2025 - Present

Designed LangChain-based conversational workflows to query unstructured patient records using HIPAA-compliant RAG with Pinecone vector databases. Built an end-to-end patient readmission prediction pipeline using PyTorch and LightGBM on a 500GB Snowflake warehouse, reducing acute hospital readmissions by 14%. Fine-tuned foundational LLMs with QLoRA to summarize extended telehealth transcripts, reducing manual physician documentation by 22%. Orchestrated CI/CD deployments for multiple production models on AWS SageMaker using Docker and Kubernetes, monitoring drift with MLflow and sustaining 0.92 AUC-ROC. Developed TensorFlow CNNs for medical imaging triage marker extraction, improving triage speed by 35%. Built secure data-masking and ELT pipelines using PySpark and Airflow for privacy and audit compliance.

Data Scientist at Aspire Technolab

June 1, 2020 - July 1, 2023

Architected distributed LSTM sequence models in PyTorch and Databricks for multi-year transaction streams to optimize supply chain allocations, reducing aggregate inventory holding costs by 16%. Implemented Isolation Forest and SVM anomaly detection on streaming telemetry ingested via Apache Kafka to stop $12K/year fraudulent financial transactions. Built collaborative filtering and XGBoost personalization engines on Google Cloud, improving conversion rates by 19% across three retail marketplaces. Applied K-Means/DBSCAN clustering and UMAP on unstructured CRM payloads to identify high-value user cohorts. Developed low-latency inference REST APIs with FastAPI and vLLM for 10,000+ daily concurrent users (<50 ms). Created statistical hypothesis testing frameworks to evaluate production rollouts of predictive pricing models for data-driven decisions.