AI Agent Developer AWS Bedrock & LLMs

About the Role

Key Responsibilities

• Design, develop, and deploy multi-step, multi-tool AI agents capable of reasoning, decision-making, and task execution.
• Leverage frameworks like Lang-Chain, Semantic Kernel, or custom-built orchestration layers to power agent behavior.
• Craft, refine, and optimize advanced prompts to improve LLM accuracy, reliability, and contextual performance.
• Develop and manage prompt chains, system prompts, and dynamic context strategies to enable adaptive agent responses.
• Continuously test and iterate prompts based on performance metrics and user feedback.
• Implement generative AI solutions using Amazon Bedrock with models from providers such as Anthropic (Claude), Cohere, Stability AI, and Amazon Titan.
• Utilize Bedrock capabilities including Guardrails, Knowledge Bases, and model customization to ensure safe and scalable AI deployments.

Requirements

• Overall IT Experience: 9-12+ Years.
• Proven experience developing autonomous AI agents using LangChain, Semantic Kernel, or similar frameworks.
• Expertise in prompt engineering, including design, chaining, and iterative refinement for large language models (LLMs).
• Hands-on experience with Amazon Bedrock (model integration, guardrails, knowledge bases).
• Proficiency in Python, API integration, and vector databases such as FAISS or Pinecone.
• Strong understanding of LLM behavior, Retrieval-Augmented Generation (RAG) architecture, and AWS deployment best practices.

Nice to Have

• Advanced knowledge of AI agent orchestration frameworks.
• Experience fine-tuning or customizing foundational models.
• Familiarity with prompt evaluation tools and automated testing for LLMs.
• Exposure to real-world AI use cases in enterprise-scale environments.

Qualifications

• No specific formal qualifications listed in the description.