[Hanoi] Principal AI/LLM Prompt Engineer

Job Description

We are seeking a Principal AI/LLM Prompt Engineer with deep expertise in Large Language Models (LLMs), prompt design, and AI system architecture.
You will lead the design, development, and operation of practical AI Agent systems that accelerate feature development in our accounting products — one of the fastest-growing SaaS solutions in Japan.

This role is both hands-on and strategic: you’ll architect scalable AI solutions, guide cross-functional teams, and shape the company’s direction in AI-driven product innovation.

Responsibilities

• Architect and build AI Agent platforms or applications from the ground up, focusing on automating and enhancing back-office operations through LLM technologies.
• Design, test, and optimize high-quality prompts, prompt-chaining strategies, and evaluation pipelines for production-grade reliability.
• Monitor and improve AI agent behavior, ensuring stability, reducing hallucinations, and optimizing performance and accuracy.
• Champion AI adoption and best practices, establishing workflows, frameworks, and cultural foundations for safe and fast-paced AI development.
• Collaborate closely with Backend, Frontend, and Infrastructure teams to ensure seamless integration of AI systems into core products.
• Mentor and coach engineers, fostering capability in prompt engineering, LLM orchestration, and AI evaluation methodologies.
• Lead AI technical initiatives across teams (BE/FE/AI/SRE), taking ownership of architecture, technical direction, and delivery quality.
• Act as a Tech Lead for AI-related projects, ensuring alignment with company strategy and technical excellence.

Requirements:

Must Have: 

• Good English skills(written and verbal).
• 4+ years of experience developing AI integrations using Python within SaaS or backend systems.
• Proven track record in prompt engineering and LLM optimization in real production environments.
• Strong understanding of web application architecture, including Frontend (TypeScript/React), Backend (web frameworks, databases), and Cloud (Azure, AWS, Terraform, serverless).
• Hands-on experience with:
  • Multi-agent orchestration and LLM frameworks (e.g., AutoGen, LangChain, LlamaIndex)
  • RAG architectures and vector databases
  • Integrating AI services such as Azure AI Foundry, AWS Bedrock, or GCP Vertex AI.

Nice to Have: 

• A background of web application development (around 5 years of experience)
• Knowledge of fine-tuning or instruction-tuning LLMs.
• Understanding of AI system deployment from a security and compliance perspective.
• Experience with CI/CD pipelines and production system operations.
• Background in B2B SaaS application development.
• Experience applying machine learning or data-driven techniques to optimize business workflows.