Description

You will design and deploy agentic AI workflows and LLM applications using AWS Bedrock and Claude, focusing on secure, loop-based coding systems.

Responsibilities

• Build agentic prompting systems with automated error-correction loops and regression testing hooks for AI-generated code.
• Develop proxy architectures to enforce security guardrails, manage API keys, and load-balance across AWS Bedrock endpoints.
• Deploy ML workloads on AWS using SageMaker, Lambda, and S3, integrating with Databricks for data processing.
• Implement MLOps practices using MLflow, Kubeflow, and model registries for model lifecycle management.
• Apply clinical NLP and healthcare data expertise to solve complex domain-specific problems.

Required Skills

• 5+ years of hands-on ML engineering in Python using PyTorch, TensorFlow, or scikit-learn.
• Deep knowledge of Claude Code customization, Agentcore, and advanced agentic prompting techniques.
• Experience with LLM architectures, RAG systems, and vector databases.
• Strong SQL skills for feature engineering and data manipulation.
• Proficiency with AWS services including SageMaker, Bedrock, Lambda, and S3.
• Experience with MLOps tooling such as MLflow or Kubeflow.
• Familiarity with Databricks and NLP libraries.

Preferred Skills

• Experience with LiteLLM gateway and cloud-based coding agent workflows.
• Background in healthcare data or clinical NLP.

Key Skills

Education

Bachelor's degree