Senior, Hands-On Engineer

About Us

We are a local construction company looking to disrupt the industry.

We are searching for a senior, hands-on engineer to design, build, secure, and operate a private AI environment in Vancouver.

The initial project is to deploy and maintain an LLM and OpenClaw on 2× NVIDIA DGX Spark or equivalent hardware, with local model serving, offline package and model mirroring, secure private interconnects, recovery media, documented runbooks and full validation.

After the initial build, this role will continue as the technical owner for the environment and its ongoing AI work, including model upgrades, performance tuning, prompt and agent workflow development, tool integrations, evaluations, safety controls and internal AI system improvements.

This is not a generic prompt-engineering role. This is a build-and-own role spanning secure infrastructure, offline deployment, local LLM serving, agent systems and ongoing applied AI development.

The ultimate goal of this role is to successfully develop, implement and manage AI deployment throughout the organization.

What You’ll Do

• Design and deploy a secure AI platform from the ground up
• Install, configure, and operate an LLM and future local/open-weight models
• Deploy and maintain OpenClaw with local model providers such as vLLM, SGLang, Ollama, or similar
• Configure and support DGX Spark or equivalent Linux/GPU systems, including storage, networking, containerization, backups, and recovery
• Build offline workflows for package mirroring , model downloads and verification, updates and rollback, backup and restore and recovery media creation
• Implement hardening for a secure environment, including network isolation, access controls, operational safeguards and logging
• Benchmark and tune model serving for context length, throughput, latency and reliability
• Create clear runbooks and acceptance tests, including proof that the system works with external internet disconnected
• Support the ongoing AI roadmap, including new models, prompt workflows, tool use, evaluations, agent behavior and internal AI automation

Required Qualifications

• 5+ years in Linux systems, DevOps, platform engineering, MLOps, AI infrastructure, or a closely related field
• Proven hands-on experience deploying local/open-weight LLMs on GPU hardware
• Strong skills in Ubuntu/Linux, Bash, Python, Docker, networking and system troubleshooting
• Experience with one or more of vLLM, SGLang, Ollama, Ray, CUDA, NCCL or similar inference/distributed tooling
• Experience working in secure, offline, or tightly controlled environments
• Strong documentation skills and the ability to own a system end-to-end

Preferred Qualifications

• Experience with NVIDIA DGX Spark, DGX systems, or AI workstations/servers
• Experience with OpenClaw or similar agent platforms
• Familiarity with Qwen, Hugging Face offline workflows, tensor parallelism and model evaluation
• Experience across both ARM64 and x86 Linux environments
• Background in security hardening, backup/recovery or regulated/private infrastructure
• Experience building practical internal AI workflows, not just experiments

Why Join Us

• Build a serious private AI platform from the ground up
• Own both the infrastructure and the applied AI layer
• Work on real-world local agent systems, not just demos
• Help define the long-term direction of a secure on-prem AI environment

Pay

From $100,000.00 per year

Benefits

• Casual dress
• Company events
• Disability insurance
• Employee assistance program
• Extended health care
• Flexible schedule
• Life insurance
• On-site parking
• Paid time off
• Vision care
• Work from home

Work Location

In person