AI Agentic Systems Intern — Real-World Impact on Azure

Zurich Insurance; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

Zurich Insurance seeks a hands-on AI engineer to design and ship agentic AI systems that perform multi-step reasoning, tool use, and autonomous decision-making in real business workflows. You will build AI-powered applications with LLMs, craft prompt strategies, and deploy pipelines using Docker and GitHub Actions.

You will integrate agents with structured and unstructured data sources, collaborating with stakeholders to re-engineer workflows for better decisions.

Qualifications

• Strong ability to drive solutions from idea to delivery.
• Experience building AI-powered applications and pipelines.
• Familiarity with integrating structured and unstructured data sources.

Responsibilities

• Design and build agentic AI systems for multi-step reasoning and tool use across real workflows.
• Develop AI-powered applications using LLMs, including prompt engineering and RAG pipelines.
• Deploy AI-driven solutions using frameworks like FastAPI, Celery, and containerized with Docker and GitHub Actions.
• Integrate agents with data sources, enabling memory and retrieval layers for reliability.
• Collaborate with business stakeholders to map workflows and re-engineer them as agentic solutions.
• Architect and scale solutions on Azure for production use.
• Monitor, evaluate, and iterate on deployed AI systems with strong observability.

Skills

Python

LLMs

Data Science

Data Wrangling

Business Communication

Education

Bachelor's or Master's in CS/Math/Data Science

Tools

Docker

GitHub Actions

Azure

Job description

Zurich Insurance seeks a hands-on AI engineer to design and ship agentic AI systems that perform multi-step reasoning, tool use, and autonomous decision-making in real business workflows. You will build AI-powered applications with LLMs, craft prompt strategies, and deploy pipelines using Docker and GitHub Actions.

You will integrate agents with structured and unstructured data sources, collaborating with stakeholders to re-engineer workflows for better decisions.