LLM Infra Production Engineer & SRE

Tribus; Hong Kong; On-site; HKD 420,000 - 660,000; Full time

Job summary

Tribus in Hong Kong is seeking a Production Support Engineer to operate and improve the LLM platform, sitting between Production Engineering, SRE and AI infrastructure. You will work across modern cloud and distributed systems to keep services reliable and responsive.

This hands-on role requires strong Linux/Windows skills, scripting proficiency and solid incident-management. You will collaborate with engineering and external providers to resolve incidents and continuously improve tooling and

Qualifications

• Experience owning production incidents.
• Experience supporting LLM infra or trading.
• Strong Linux and Windows knowledge.
• Proficient in Python, Bash or PowerShell.
• Experience with AWS, Kubernetes and distributed systems.
• Experience with monitoring/observability (Grafana/Prometheus).
• SQL experience with PostgreSQL or SQL Server.
• Excellent communication and incident-management skills.

Responsibilities

• Own and troubleshoot production issues across the LLM platform and model-provider integrations.
• Investigate latency, availability, API, capacity and regional performance issues.
• Support AWS/Kubernetes infrastructure and developer tooling.
• Build scripting and automation to improve diagnostics and reduce manual work.
• Improve monitoring, dashboards and alerting.
• Support releases and production changes.
• Work with engineering, infrastructure teams and external providers to resolve incidents.

Skills

Production Support

SRE

Platform Ops

LLM/AI infra

Trading funds

Linux

Windows

Python

Bash

PowerShell

AWS

Kubernetes

Grafana

Prometheus

SQL

PostgreSQL

SQL Server

Communication

Incident Mgmt

Tools

Grafana

Prometheus

Job description

Tribus in Hong Kong is seeking a Production Support Engineer to operate and improve the LLM platform, sitting between Production Engineering, SRE and AI infrastructure. You will work across modern cloud and distributed systems to keep services reliable and responsive.

This hands-on role requires strong Linux/Windows skills, scripting proficiency and solid incident-management. You will collaborate with engineering and external providers to resolve incidents and continuously improve tooling and