Solutions Engineer; Lead

Solutions Engineer (Lead)

Location:

Belfast, Northern Ireland, United Kingdom

Hours:

Full-time

Salary: £70,000 - £85,000 per annum (plus EMI share options)

Whitespace leads in creating smarter, safer AI solutions for enterprises. As AI presents transformative opportunities, many companies struggle to harness its full potential. Whitespace bridges this gap with Collective – The AI Operating System for high‑assurance environments. Built for scale, trust and action – even in the most secure, complex or disconnected settings. We're not just building AI systems. We're laying the foundation for a future where humans and machines work together like never before.

The successful AI Solutions Engineer will extend and enhance our AI Operating System, which leverages LLMs to solve industry‑specific challenges across defence, legal, health, infrastructure and management consulting sectors. This is a hands‑on lead role focused on rapidly prototyping and deploying AI‑powered solutions. Working directly with clients, you will translate their needs into scalable, production‑ready AI applications using modern frameworks and techniques.

• Develop platform functionality using Python, building APIs and integrations to extend capabilities for diverse client needs.
• Design and implement LLM‑powered applications and workflows using open‑source models such as Llama, Qwen and Gemma, as well as models from OpenAI, Gemini, etc.
• Build AI agents with tool/function calling, prompt engineering and appropriate guardrails using frameworks such as OpenAI Agent
  
  SDK, Lang Graph or Llama Index.
• Implement testing and evaluation frameworks for LLM applications, covering prompt testing, output quality metrics and agent behaviour validation.
• Apply relevant AI technologies as needed, including retrieval systems (RAG, Graph
  
  RAG), knowledge graphs, vector databases or data pipelines.

• Design and deliver scalable, efficient solutions tailored to client‑specific requirements whilst balancing innovation with practical constraints around compliance, performance, cost and security.
• Lead technical discussions, diagnose complex challenges and provide guidance to less experienced team members.

• Work directly with clients to gather requirements and translate them into actionable AI‑powered solutions, moving rapidly from concept to working prototype.
• Communicate complex technical concepts effectively to technical and non‑technical stakeholders, including system capabilities and limitations.
• Mentor junior engineers and contribute to technical knowledge sharing within the team.

• At least five years as a software engineer on commercial platforms, with demonstrable experience building production‑ready LLM‑powered applications.
• Proven experience with API‑level LLM usage, including tool/function calling, prompt engineering and evaluation.
• Experience with agent frameworks (OpenAI Agent
  
  SDK, Lang Graph, Llama Index Agents or similar).
• Experience developing APIs using FastAPI or similar frameworks and integrating with third‑party platforms.
• Direct client‑facing experience gathering requirements and delivering technical implementations.
• Experience with in agile development workflows and engineering teams.

• Strong Python (or similar) programming skills with a focus on production‑grade applications.
• Excellent communication abilities, translating complex technical concepts for diverse audiences.
• Strong analytical and problem‑solving approach, identifying scalable and reusable solutions.
• Leadership qualities, including technical mentorship, team collaboration and line management.
• Ability to align solutions with business goals and industry‑specific constraints.
• Self‑sufficient contributor capable of working independently and seeking support when needed.

• Open source LLMs (Llama, Qwen, Gemma, GPT‑OSS) and local deployment strategies.
• Frameworks and protocols such as Model Context Protocol (MCP) or Agent‑to‑Agent (A2A).
• LLM evaluation tooling (OpenAI Evals, Lang Smith, custom evaluation harnesses).
• Advanced agent patterns: multi‑agent…