Production LLM Engineer: Scalable AI, RAG & Guardrails

HRBS Global; Islamabad; On-site; PKR 1,800,000 - 3,200,000; Full time

Job summary

HRBS Global is hiring an experienced AI/LLM Engineer to build and optimise a production-grade, customer-facing AI assistant for a UK-based client. Location: Islamabad, on-site with UK hours (12 PM–9 PM PKT).

You will design scalable LLM systems, develop RAG pipelines, embeddings, fine-tuning workflows, and guardrails, collaborating with compliance and legal teams to ensure reliable, observable, and auditable AI solutions.

Qualifications

• Production experience with Python and FastAPI.
• Experience deploying LLM applications in production.
• Hands-on with RAG, embeddings, and vector databases.
• Prompt engineering and AI evaluation experience.
• Familiar with Docker, Redis, AWS, APIs, and databases.
• Knowledge of LLM performance optimization and scaling.
• Open‑source LLMs or GPU deployment is a plus.

Responsibilities

• Design and deploy scalable LLM systems for production use.
• Build and optimise RAG pipelines and knowledge retrieval systems.
• Develop embedding pipelines and work with vector databases.
• Implement fine-tuning workflows and model evaluation pipelines.
• Develop AI guardrails, policy enforcement, and escalation mechanisms.
• Optimise model performance, latency, concurrency, and GPU utilisation.
• Ensure AI systems are reliable, observable, and fully auditable.
• Integrate AI solutions with databases, APIs, and existing applications.
• Collaborate with technical, compliance, and legal teams.

Skills

Python

LLM engineering

Prompt engineering

AI evaluation

System design

Observability

Compliance awareness

Tools

Docker

Redis

AWS

APIs

Databases

Vector databases

vLLM

GPU deployment

FastAPI

Job description

HRBS Global is hiring an experienced AI/LLM Engineer to build and optimise a production-grade, customer-facing AI assistant for a UK-based client. Location: Islamabad, on-site with UK hours (12 PM–9 PM PKT).

You will design scalable LLM systems, develop RAG pipelines, embeddings, fine-tuning workflows, and guardrails, collaborating with compliance and legal teams to ensure reliable, observable, and auditable AI solutions.