AI Agent Developer

IntelliPro; Hong Kong; On-site; HKD 900,000 - 1,500,000; Full time

Job summary

IntelliPro seeks an experienced AI Agent Developer to design and deliver production-grade autonomous AI Agents capable of task understanding, planning, and tool orchestration. You will architect multi-agent systems with memory management and context engineering, and implement agent frameworks for real-world business scenarios.

The role requires strong Python skills, hands-on experience with major agent frameworks, and a track record of integrating LLMs via APIs, managing prompts and model

Qualifications

• 3+ years of hands-on experience building AI Agent systems in production or near-production environments.
• Deep understanding of core Agent paradigms: Multi-Agent orchestration, Planning & Reasoning, Tool Use, Memory, Context Engineering.
• Proven track record with at least one major Agent framework: LangGraph, AutoGen, Dify, Coze (or equivalent).
• Strong programming skills in Python; familiarity with async/concurrent execution patterns for agent workflows.
• Experience integrating LLMs (GPT, Claude, Gemini, or open-source models) via APIs and managing prompt strategies, token budgets, and model routing.
• Solid grasp of RAG, vector databases, embedding strategies, and retrieval-augmented agent workflows.
• Experience building systems that interact with external tools, APIs, and databases in a safe, auditable manner.
• Ability to design agent evaluation metrics and run systematic experiments to improve agent performance.
• Strong communication skills; comfortable explaining agent behavior, failure modes, and trade-offs to both technical and non-technical stakeholders.

Responsibilities

• Design, build, and deploy autonomous AI Agents capable of task understanding, reasoning, planning, and tool orchestration.
• Architect Multi-Agent systems with structured collaboration patterns, memory management, and context engineering.
• Implement and optimize Agent frameworks (LangGraph, AutoGen, Dify, Coze, or similar) for real-world business scenarios.
• Develop robust Tool Use layers — defining tool schemas, managing tool routing, handling error recovery, and ensuring safe execution.
• Engineer long-term and short-term memory architectures to support stateful, context-aware agent behaviors.
• Design reasoning pipelines (chain-of-thought, tree-of-thought, ReAct, etc.) and integrate them into production workflows.
• Collaborate with domain experts to map complex business processes into agent-executable plans.
• Establish evaluation frameworks to measure agent reliability, latency, accuracy, and cost efficiency.
• Stay current with the evolving Agent landscape and contribute to internal best practices and architecture decisions.

Skills

Python programming

Async/concurrent execution

LLMs integration

Multi-Agent orchestration

Communication skills

Tools

LangGraph

AutoGen

Dify

Coze

Job description

We are building autonomous AI Agents that understand complex tasks, plan multi-step workflows, invoke tools, and deliver end-to-end results across domains — including trading analysis, customer service augmentation, risk investigation, and operations automation. We are looking for an experienced AI Agent Developer to drive the design and delivery of these production-grade agentic systems.

Responsibilities

• Design, build, and deploy autonomous AI Agents capable of task understanding, reasoning, planning, and tool orchestration.
• Architect Multi-Agent systems with structured collaboration patterns, memory management, and context engineering.
• Implement and optimize Agent frameworks (LangGraph, AutoGen, Dify, Coze, or similar) for real-world business scenarios.
• Develop robust Tool Use layers — defining tool schemas, managing tool routing, handling error recovery, and ensuring safe execution.
• Engineer long-term and short-term memory architectures to support stateful, context-aware agent behaviors.
• Design reasoning pipelines (chain-of-thought, tree-of-thought, ReAct, etc.) and integrate them into production workflows.
• Collaborate with domain experts to map complex business processes (trading analysis, risk investigation, customer support) into agent-executable plans.
• Establish evaluation frameworks to measure agent reliability, latency, accuracy, and cost efficiency.
• Stay current with the evolving Agent landscape and contribute to internal best practices and architecture decisions.

Requirements

• 3+ years of hands-on experience building AI Agent systems in production or near-production environments.
• Deep understanding of core Agent paradigms: Multi-Agent orchestration, Planning & Reasoning, Tool Use, Memory, Context Engineering.
• Proven track record with at least one major Agent framework: LangGraph, AutoGen, Dify, Coze (or equivalent).
• Strong programming skills in Python; familiarity with async/concurrent execution patterns for agent workflows.
• Experience integrating LLMs (GPT, Claude, Gemini, or open-source models) via APIs and managing prompt strategies, token budgets, and model routing.
• Solid grasp of RAG, vector databases, embedding strategies, and retrieval-augmented agent workflows.
• Experience building systems that interact with external tools, APIs, and databases in a safe, auditable manner.
• Ability to design agent evaluation metrics and run systematic experiments to improve agent performance.
• Strong communication skills; comfortable explaining agent behavior, failure modes, and trade-offs to both technical and non-technical stakeholders.