AI Systems Architect at UltaHost — Remote AI Engineering and Infrastructure Role

About the role

UltaHost is a global web hosting and cloud infrastructure provider offering VPS hosting, dedicated servers, shared hosting, domains, game servers and cloud-based services.

As the company expands its use of artificial intelligence, it is seeking an AI Systems Architect to help build the technical foundation for AI-powered products, internal platforms and business automation.

This is a hands-on technical role, not a traditional AI research position. The successful candidate will combine infrastructure engineering, AI application development and product thinking to identify practical problems and develop reliable solutions that deliver measurable business value.

You will work directly with the Chief Technology Officer (CTO) and collaborate with Product, Engineering, Infrastructure, Operations, Support, Sales, Marketing and other departments.

The position is an individual-contributor role with significant technical ownership and architectural influence. Direct team-management responsibilities are not currently included.

The ideal candidate can move across the production stack, from GPU infrastructure and virtualisation to language-model deployment, APIs, AI agents, security, monitoring and business workflow automation.

Responsibilities

1. AI and GPU infrastructure

• Design, deploy and operate infrastructure for AI workloads and large language models (LLMs) on GPU-enabled servers.
• Evaluate GPU requirements, memory capacity, model size, quantisation, throughput, latency and operating costs.
• Design infrastructure for virtual machines, containers and secure resource isolation.
• Support GPU and PCIe passthrough where appropriate.
• Improve infrastructure capacity planning, monitoring, availability, backup and disaster recovery.
• Develop repeatable deployment processes for production AI environments.

2. Virtualisation and platform engineering

• Build and maintain Proxmox-based infrastructure for AI and general-purpose workloads.
• Work with Proxmox clusters, KVM, virtual machines, Linux containers, networking, storage and backups.
• Automate virtual machine, container and infrastructure provisioning.
• Troubleshoot performance, networking, storage and hardware-resource issues.
• Decide when workloads should run on bare metal, virtual machines, containers or orchestration platforms.

3. LLM deployment and optimisation

• Deploy and operate open-source language models on private or self-hosted infrastructure.
• Evaluate tools such as vLLM, Ollama, LiteLLM, Hugging Face and Text Generation Inference.
• Benchmark models for quality, performance, memory usage, concurrency and cost.
• Apply inference optimisation techniques such as quantisation, batching, caching and model routing.
• Compare self-hosted models with commercial APIs such as OpenAI, Anthropic Claude and Google Gemini.
• Build reusable model gateways and APIs with appropriate retries, rate limits, timeouts and fallback mechanisms.

4. AI applications, RAG and agentic systems

• Build production-ready AI applications, assistants, copilots and agents.
• Develop retrieval-augmented generation (RAG) systems using embeddings, vector databases, metadata filtering and retrieval evaluation.
• Connect AI assistants to authorised company documentation, product information, support content and operational procedures.
• Develop agents that use controlled tools, APIs, databases and internal services.
• Implement structured outputs, validation, tool calling, memory and context management.
• Integrate AI capabilities into customer-facing products where appropriate.
• Use traditional software and deterministic automation when these are more reliable than AI.

5. Backend development and integration

• Build reusable AI services, APIs and platform components.
• Integrate AI systems with company portals, billing systems, databases, support platforms, monitoring tools and internal applications.
• Work with REST APIs, webhooks, queues, caches, databases, scheduled jobs and event-driven workflows.
• Use Python, TypeScript/Node.js or other suitable technologies.
• Turn successful prototypes into tested, documented and maintainable production software.

6. Infrastructure and workflow automation

• Automate infrastructure provisioning, configuration, deployment and maintenance.
• Work with tools such as Terraform, Ansible, GitHub Actions, GitLab CI/CD and n8n.
• Identify repetitive work that can be reduced through AI-powered workflows or conventional automation.
• Build tools connecting infrastructure, engineering, support and business systems.
• Monitor AI applications for performance problems, model failures, infrastructure issues and abnormal behaviour.

7. Security and AI governance

• Apply least-privilege principles across infrastructure, applications, services and AI agents.
• Secure credentials, API keys, SSH access, service accounts and model-provider integrations.
• Implement permissions and isolation boundaries for AI applications and customer workloads.
• Protect customer data from unauthorised access across tenants and systems.
• Build controlled interfaces for agents that interact with servers, APIs and sensitive resources.
• Use approval steps, allowlists, validation, audit trails and rollback controls for sensitive operations.
• Address risks such as prompt injection, data leakage, hallucinations and unauthorised tool use.

8. Business automation and AI adoption

• Work with Support, Sales, Marketing, Finance, HR, Operations and Engineering to identify inefficient workflows.
• Design AI agents, scripts, dashboards and internal tools that reduce manual workload.
• Integrate AI into existing business processes rather than creating disconnected demonstrations.
• Evaluate new AI models, APIs and open-source technologies for practical business use.
• Document solutions and help teams understand and adopt AI systems.

Requirements

Applicants should have a strong hands-on technical background and experience taking solutions from initial concept through deployment and production operation.

Infrastructure and systems

• At least five years of experience in infrastructure engineering, platform engineering, DevOps, site reliability engineering, cloud engineering, backend systems, virtualisation or production systems administration.
• Strong Linux administration and troubleshooting skills.
• Practical experience with KVM-based virtualisation or comparable enterprise technologies.
• Experience with Proxmox VE or a closely related virtualisation platform.
• Strong Docker and containerisation knowledge.
• Understanding of virtual machines, networking, storage, snapshots, backups and resource isolation.
• Knowledge of TCP/IP, DNS, firewalls, routing, reverse proxies, TLS and service connectivity.
• Experience operating or troubleshooting production infrastructure.

AI and LLM engineering

• At least 2–3 years of practical experience building, deploying, operating or integrating AI applications, LLM systems or AI infrastructure.
• Practical understanding of generative AI, RAG, embeddings, vector search, tool calling, agents, prompts and context management.
• Experience building and deploying real AI-powered applications or systems.
• Experience with commercial LLM APIs and AI development platforms.
• Familiarity with open-source language-model ecosystems and self-hosted deployment.
• Understanding of LLM evaluation, structured-output validation, hallucination management, reliability, latency and cost optimisation.
• Ability to determine when AI is appropriate and when conventional software or automation is the better choice.

Software development and integration

• Strong programming or scripting experience with Python and/or JavaScript, TypeScript or Node.js.
• Experience integrating APIs, webhooks, databases, backend services and third-party systems.
• Familiarity with PostgreSQL, MySQL/MariaDB, Redis or vector databases.
• Experience designing reliable backend workflows and automation.
• Working knowledge of Git and modern deployment practices.
• Ability to create clean user interfaces or collaborate effectively with designers.

Production engineering and professional skills

• Experience with monitoring, centralised logging, alerting, health checks and production troubleshooting.
• Understanding of high availability, backup, failure recovery and disaster recovery.
• Strong debugging and root-cause-analysis skills.
• Knowledge of security principles, including least privilege, credential management and system isolation.
• Strong analytical, problem-solving and communication skills.
• Professional working proficiency in English.
• Ability to work independently in a fully remote environment.
• Experience taking technical solutions from concept or prototype into production.

Preferred qualifications

Experience in several of the following areas would be an advantage:

• Web hosting, VPS, dedicated servers, cloud infrastructure or data centres.
• Proxmox VE, Proxmox Backup Server, KVM and GPU passthrough.
• ZFS, Ceph or distributed storage.
• NVIDIA GPU infrastructure and CUDA.
• vLLM, Ollama, LiteLLM or NVIDIA Triton.
• LLM quantisation, model customisation or multi-GPU inference.
• LangChain, LangGraph, LlamaIndex, CrewAI or similar frameworks.
• Model Context Protocol (MCP) and tool-based AI integrations.
• Qdrant, pgvector, Weaviate, Milvus or Pinecone.
• Terraform, Ansible, Kubernetes and CI/CD tools.
• n8n, Make, Redis and message queues.
• Prometheus, Grafana, OpenTelemetry, Sentry or secrets-management platforms.
• WHMCS, cPanel/WHM, HostBill, Blesta, billing systems or hosting automation.
• Infrastructure security, incident response and internal developer platforms.

Candidates do not need to be experts in every technology listed, but should understand how the components of a production AI system work together.

Benefits

The job description does not specify a salary or detailed employee benefits package.

The role offers the opportunity to:

• Work directly with the CTO on the company's AI strategy and technical architecture.
• Build AI-powered infrastructure, applications and automation used across an international hosting business.
• Work across GPU infrastructure, self-hosted LLMs, backend engineering and intelligent automation.
• Influence architecture and technology decisions without direct team-management responsibilities.
• Develop production AI systems rather than limiting work to prototypes and experiments.
• Work remotely with teams across engineering, infrastructure, product and business operations.

Salary, paid leave, insurance, equipment allowances, relocation support and other employment benefits should be confirmed with UltaHost.

How to apply

1. Visit the official UltaHost vacancy page: https://ultahost.keka.com/careers/jobdetails/7730
2. Review the job requirements and confirm your eligibility for the remote position.
3. Prepare an up-to-date CV highlighting infrastructure engineering, Linux, virtualisation, backend systems and practical AI experience.
4. Showcase projects you have built or deployed, particularly production AI applications, RAG systems, AI agents, self-hosted language models or infrastructure automation.
5. Explain your personal contribution, the technical challenges you solved and the measurable results of your work.
6. Highlight relevant experience with GPU infrastructure, Proxmox, KVM, Docker, cloud systems, LLM APIs and automation tools.
7. Submit your application through the official recruitment portal and follow the instructions provided.

Application link: https://ultahost.keka.com/careers/jobdetails/7730

Important for applicants in Nigeria and elsewhere in Africa: The vacancy states that the role is fully remote, but it does not specify which countries are eligible. Confirm geographic eligibility, salary arrangements and any employment or work-authorisation requirements directly with UltaHost before applying.