Senior AI Inference Engineer - High-Performance Serving

Aleroy; United Arab Emirates; On-site; AED 360,000 - 540,000; Full time

Benefits offered by this job

24 days annual leave

Health insurance

Modern office in Yas Creative Hub

Job summary

Aleroy aims to accelerate AI model deployment with a unified platform. This senior role focuses on building and optimising the inference serving stack, driven by hands-on work on the stack and measurement-driven improvements.

You will lead performance engineering across GPUs, MoE models, and distributed serving, setting standards and guiding engineers with a practical, outcome-oriented approach. The role is based in our Yas Creative Hub office in the UAE.

Qualifications

• Strong hands-on experience with LLM inference-serving in production.
• Deep knowledge of vLLM, SGLang, TensorRT-LLM.
• Experience with MoE models including expert parallelism and all-to-all communication.
• GPU performance engineering using CUDA or Triton.
• Experience with model quantization and performance optimisation.
• Strong benchmarking and measurement discipline across serving performance and cost.
• English level: Advanced (C1).

Responsibilities

• Own and improve the FAR Labs inference serving stack with optimisations.
• Improve throughput, latency, memory efficiency, and serving cost while maintaining model-quality standards.
• Identify and remove GPU performance bottlenecks across the stack using profiling and optimization.
• Drive efficient MoE serving including expert parallelism and load balancing.
• Implement and evaluate quantization and other model-optimisation techniques.
• Own rigorous, reproducible benchmarking across serving performance and cost.
• Build instrumentation to understand system behaviour and validate improvements.
• Make technical decisions around routing, scheduling, and distributed serving.

Skills

LLM serving

Performance tuning

Distributed systems

CUDA/Triton

MoE models

Quantization

Benchmarking

Leadership

Instrumentation

English (C1)

Tools

vLLM

SGLang

TensorRT-LLM

CUDA

Triton

Job description

Aleroy aims to accelerate AI model deployment with a unified platform. This senior role focuses on building and optimising the inference serving stack, driven by hands-on work on the stack and measurement-driven improvements.

You will lead performance engineering across GPUs, MoE models, and distributed serving, setting standards and guiding engineers with a practical, outcome-oriented approach. The role is based in our Yas Creative Hub office in the UAE.