About GHD

GHD is committed to creating positive impact through our work, supporting our people with the technology, training, and opportunities to build meaningful outcomes for our clients and communities.

The opportunity

We are looking for a hands-on AI Engineer to help GHD design and deliver production-grade GenAI capabilities for internal teams. You will build reusable components, reference patterns, and evaluation tooling that enable both software engineering teams (pro-code) and business enablement teams (low-code) to embed AI safely into products and internal platforms.

This is an internal role focused on building scalable, maintainable AI capabilities across the business. It is not a client-facing consulting position.

You will work closely with architects, product owners, and delivery teams to move GenAI from pilots to production with strong quality, governance, and operational practices.

You will thrive in this role if you are a senior software engineer who has recently built GenAI features in production, and you enjoy turning one-off solutions into reusable patterns, templates, and quality checks that other teams can adopt safely.

What you will do

• Build and maintain GenAI application components such as agent templates, tool/function calling patterns, orchestration flows, and reusable services.
• Develop RAG and context management patterns (retrieval strategies, chunking, grounding, citations, memory boundaries, prompt structure).
• Create and run evaluation and quality frameworks including automated test suites, regression checks, human review workflows, and performance/safety metrics.
• Enable adoption across teams by producing reference implementations, documentation, and developer guidance that work in both pro-code and low-code contexts.
• Partner with stakeholders to identify high-value internal use cases and turn them into deliverable, supportable solutions.
• Ensure solutions align to enterprise expectations for security, privacy, reliability, logging/monitoring, and support.
• Contribute to standards, playbooks, and governance processes for responsible and safe AI.
• Support incident and problem management for AI-enabled services when required, including triage and root cause analysis.

What we are looking for (must have)

• Strong software engineering background with C# or Python (ideally both) and experience building and supporting production applications.
• Demonstrated experience delivering at least one LLM-enabled application or agent beyond a demo (examples: tool calling, RAG, workflow orchestration, guardrails, or production rollout).
• Experience with the Microsoft cloud ecosystem for modern application delivery (Azure services, identity, security, monitoring, CI/CD).
• Practical understanding of prompting patterns, context management, reliability techniques, and common GenAI failure modes (hallucination, prompt injection, data leakage, brittle tool use).
• Ability to work with multiple teams and stakeholders, explain trade-offs clearly, and drive alignment on standards and patterns.
• Strong documentation and communication skills.

Nice to have

• Hands-on experience with Azure OpenAI and one or more orchestration frameworks (for example Semantic Kernel, AutoGen, LangGraph, LangChain).
• Experience designing evaluation frameworks (test datasets, automated scoring, red-teaming approaches, safety checks).
• Experience integrating GenAI into web apps or APIs (Angular, React, HTML5, REST).
• Familiarity with interoperability standards and agent protocols (for example MCP, A2A).
• Experience working alongside low-code teams and building safe “guardrailed” pathways to enterprise services.

Why join

• Build real, production GenAI capabilities that are used across the organisation.
• Work in a Microsoft-first environment with strong focus on responsible AI and engineering quality.
• Opportunity to shape reusable patterns and standards that uplift multiple teams, not just one project.
   

We embrace hybrid work arrangements, with an expectation of being in the office for 3 days each week.