Regional Agentic AI Engineer — Build Enterprise LLM Runtimes

Argyll Scott

Hong Kong

Hybrid

HKD 1,000,000 - 1,300,000

Full time

3 days ago

Be an early applicant

Get a reply from this recruiter

Your resume and cover letter, tailored to exactly what they’re hiring for.

Get past ATS filters

Job summary

Argyll Scott is partnering with a global life insurer to build out its regional AI Centre of Excellence in Hong Kong, with hybrid options for Asia-Pacific deployments. As an Agentic AI Engineer, you will develop the runtimes that empower agents to plan, execute and recover complex work, blending orchestration, tool integration and production operations.

You’ll work closely with AI Quality & Evals Engineers to make agentic systems dependable, observable and enterprise-ready, leveraging AWS

Qualifications

5+ years delivering production software, including 2+ years building LLM-powered systems in production.
Strong Python skills and asynchronous programming capabilities.
Hands-on experience with retrieval engineering across embeddings and vector search.
Knowledge of data governance, catalogues or semantic-layer systems.

Responsibilities

Build durable agent runtimes with planner–executor loops and recovery logic.
Implement human approval gates and safe handling of high-impact actions.
Standardise tool integration via APIs and MCP patterns.
Develop deterministic replay to test agent decisions and tool calls.
Connect agents to metadata, data catalogs and semantic layers with access controls.
Engineer RAG pipelines spanning retrieval, reranking and source attribution.
Create evaluation suites, deployment gates and runtime monitoring for quality and cost.