about 3 hours ago
Responsibilities
- Design, build, and operate multi-agent workflows and tool-enabled agents.
- Architect and maintain end-to-end RAG systems for document processing.
- Evaluate and integrate LLMs and GenAI services based on cost and performance.
- Develop and optimize prompting strategies with automated testing.
- Define evaluation frameworks for generative outputs and monitor performance.
- Apply classical NLP techniques and manage data distribution shifts.
- Build and operate scalable AI infrastructure on AWS.
- Own the full deployment lifecycle including CI/CD and testing strategies.
- Ensure data quality through validation and proactive dataset sourcing.
Requirements
- Production experience designing multi-agent systems with orchestration logic.
- Hands-on experience building RAG pipelines end-to-end.
- Strong experience defining and running evaluation pipelines for generative AI.
- Demonstrated experience with foundation models and GenAI providers.
- Solid grounding in classical NLP techniques and their applications.
- Hands-on experience with Amazon Bedrock and its APIs.
- Proficiency with AWS services including SageMaker and Lambda.
- Familiarity with MLOps practices and CI/CD for ML.
- Experience with Infrastructure as Code using Terraform or AWS CDK.
- Production-quality Python skills including testing and clean ML abstractions.
- Familiarity with traditional ML frameworks and fine-tuning workflows.
- Experience with experiment tracking tools for metrics and artifacts.
- Ability to translate business goals into technical solutions.
- Strong collaborative instincts across engineering and product teams.
- A rigorous, evidence-driven mindset for shipping confident solutions.