AI Engineer

a. Design and build a reusable multi-agent AI system that can support multiple projects and use cases

b. Define agent roles, responsibilities, and interaction patterns, including planning, coordination, and handoff

c. Implement agent orchestration using frameworks such as: LangChain, LangGraph, CrewAI

d. Integrate both self-hosted LLM models and commercial LLM services

e. Develop backend services in Python, including: Agent orchestration and execution logic, Tool integration and external API access, Backend APIs for frontend consumption

f. Develop and maintain frontend components using React.js, including: User interfaces for interacting with AI agents, Visualisation of agent outputs and intermediate results, Human-in-the-loop workflows such as review, approval, and override

g. Implement agent evaluation and monitoring mechanisms, covering: Response quality and correctness, Performance and latency, Cost and resource usage

h. Work with business users from different departments to understand requirements and workflows, and translate business problems into agent-based solutions

i. Produce technical documentation, including design documentation, agent patterns and usage guidelines

The suitable candidates ideally need to meet the following requirements:

a. Strong hands-on experience with multi-agent AI system design and implementation

b. Proficient in Python for backend development and React.js for frontend development

c. Practical experience using LangChain, LangGraph, CrewAI, or similar agent frameworks

d. Experience working with both self-hosted LLMs and commercial LLM APIs

e. Familiarity with agent evaluation approaches and AI quality assessment

f. Good understanding of LLM limitations, such as hallucinations and non-deterministic behaviour

g. Ability to design reliable systems with appropriate guardrails and human-in-the-loop controls

h. Strong communication skills and ability to engage business users with varying technical backgrounds

i. Able to work independently while collaborating closely with internal teams