We are looking for an experienced AI Engineer with strong expertise in Python, AWS, data engineering, RAG, vector search, and agentic AI solutions.
Key Responsibilities:
- Build and deploy AI/data pipeline containers using Docker, ECR, AWS Batch, EKS, and Databricks.
- Develop RAG and retrieval workflows using Chroma, OpenSearch, Amazon Kendra, Bedrock Embeddings, and LangChain.
- Design and implement agentic AI solutions, including single-agent, workflow, and multi-agent architectures.
- Build agent orchestration for task decomposition, routing, tool use, context/state management, retries, and human approval.
- Integrate MCP servers with appropriate security, permissions, tool discovery, and error handling.
- Implement quality, safety, groundedness, and output-validation checks for AI workflows.
- Build document ingestion and processing pipelines using AWS Textract, OCR, Python, Pandas, and PySpark.
- Work with Databricks, Delta Lake, MLflow, Unity Catalog, and PySpark where applicable.
- Apply AWS security and data-governance practices including IAM, KMS, and Secrets Manager.
- Develop automated tests and data-quality checks for pipelines.
Required Skills:
- 5+ years of professional data engineering / AI engineering experience.
- Strong Python and SQL skills.
- Experience with AWS S3 and at least one of Redshift, PostgreSQL, or DynamoDB.
- Experience with Docker, ECR, and AWS Batch.
- Knowledge of document processing, OCR, PDF/Word/Excel processing.
- Understanding of embeddings, vector databases, and retrieval systems.
- Experience with Chroma, OpenSearch, or similar technologies.
- Familiarity with IAM, KMS, Secrets Manager, and secure data handling.
- Experience building production-quality data/AI pipelines.
Preferred Skills:
- Databricks, Delta Lake, MLflow, Unity Catalog, and PySpark.
- Amazon Bedrock, Bedrock Embeddings, LangChain, or Amazon Kendra.
- AWS Glue, Athena, Step Functions, EventBridge, Kafka/MSK, or Kinesis.
- Experience with agentic AI, MCP, and multi-agent workflows.
- Terraform/IaC familiarity.
- Experience with GxP, HIPAA, GDPR, or clinical-trial data.