Description

You will design, build, and optimize multi-agent LLM architectures for chatbot orchestration. This role covers the full lifecycle from prompt engineering and RAG implementation to LLMOps and deployment on AWS.

This role is remote.

Responsibilities

• Design and implement advanced prompt chains, retrieval-augmented generation (RAG), and agent memory systems.
• Integrate multiple LLMs (OpenAI, Anthropic) and tools (vector DBs, external APIs, code execution) into cohesive agentic pipelines.
• Build and maintain LLMOps workflows using GitLab pipelines and Terraform/CloudFormation on AWS.
• Monitor, test, and evaluate LLM performance in production, defining guardrails, fallback strategies, and observability metrics.
• Mentor junior engineers, review code, and collaborate with stakeholders to translate product needs into LLM-driven solutions.

Required Skills

• 4-8 years in software or machine learning engineering, with 2+ years focused on LLMs or NLP systems.
• Deep experience with prompt engineering, LLM APIs, and frameworks like LangGraph, LangChain, or LlamaIndex.
• Proven skills in deploying and managing AI/ML workloads on AWS (SageMaker, ECS, EKS, S3, Lambda, CloudWatch).
• Strong Python coding skills; ability to write clean code independently.
• Experience with Agentic AI patterns, including LLM workflows, Agent loops, and ReAct Agents.
• Familiarity with vector databases and LLMOps tooling (LangSmith, Langfuse, etc.).
• Experience working in a product environment with Agile methodologies.

Preferred Skills

• Experience with DevOps practices and CI/CD pipelines using GitLab.
• Ability to stay ahead of LLM research trends and translate insights into engineering designs.

Key Skills

Education

Bachelor's degree