Ideally 10+ years in software engineering/infrastructure, including 5+ years deploying AI/LLM models into production environments.
Proven experience deploying AI/LLM models in on-premise, air-gapped, sovereign environments
Expertise with LLM deployment runtimes such as vLLM, TGI, Ollama or equivalents
GPU cluster management experience (NVIDIA H100/H200, NVLink topology, InfiniBand networking)
Experience applying quantization techniques (GPTQ, AWQ, GGUF) in production
Kubernetes-based model serving at scale
Familiarity with open-weight models (Kimi, DeepSeek, Qwen, LLaMA family)
Ability to design and implement deployment architecture for secure, air-gapped environments

Background in high-performance computing or distributed systems beyond AI workloads
Experience deploying models for government, intelligence, or other highly regulated sectors
Familiarity with FIPS-compliant environments and secure model packaging
Experience with model fine-tuning pipelines and LoRA adapter management
Prior work in vulnerability management or secure enterprise software environments

Initially fully remote with willingness to travel to Abu Dhabi as needed
Long-term relocation to Abu Dhabi welcomed but not mandatory
Candidates should ideally work close to Gulf Standard Time to reduce time zone challenges

Strong technical alignment prioritized over perfect fit; candidates meeting ~50-75% of profile encouraged
Candidate quality significantly more important than quantity
Confidential role; applications via referral or selected agencies only
Salary flexible; client willing to exceed budget for exceptional candidates