RightShip's engineering organisation is at an inflection point. We are moving from a team that uses AI as a productivity aid into one where agentic and autonomous systems do real delivery work, reviewed and directed by engineers who act as orchestrators. This role is the architect and engine of that shift. The Systems Agentic Engineer will build and own the backbone that makes it real: the frameworks, the integrations, the standards, the internal tooling, and the agent catalogue that, together, turn a collection of AI-assisted individuals into a genuinely AI-native engineering organisation. If you want to define what modern engineering looks like at a company committed to being world-class at it, this is that role.

The scope goes beyond individual agents. This role owns the transformation of our full software delivery lifecycle, codifying AI-first practices at every stage from grooming through to release. It also leads the formation of RightShip's Internal Developer Platform (IDP): the shared infrastructure, tooling, and automation pipelines that make every squad faster and more capable. That includes infrastructure operations pipelines that reduce manual toil at the platform level. The person in this role will not just build agents; they will architect the delivery system the whole engineering organisation runs on.

Major Responsibilities

• Design, build and operate multi-step agentic pipelines that create virtual delivery capacity, surfacing work for engineers to review, accept, or reject.
• Build and maintain the agent types that make up RightShip's virtual delivery team, including but not limited to: developer agent, security review agent, regression test-plan agent, grooming agent, and criteria cross-check agent, and expand the catalogue as new automation opportunities are identified.
• Develop AI-powered product features and tooling in Python, contributing directly to the build alongside squads rather than purely in an advisory or enablement capacity.
• Evaluate every workflow step and make an explicit decision about whether it calls for deterministic logic (rules, conditions, structured code) or a non-deterministic AI model, documenting the rationale so the team can audit and improve it over time.
• Integrate AI-first automation into the full SDLC: code generation and review, test-case creation, PR triage, dependency management, release notes, and incident summarisation.
• Act as an enabler across engineering squads, partnering with engineers and squad leads to help each team identify, scope, and own their own agentic workflows with the right guidance and tooling.
• Develop and share evaluation patterns, reference implementations, and best-practice guardrails that individual engineers and squads can adopt and adapt, keeping ownership of agents with the teams that run them.
• Evaluate and onboard orchestration frameworks (such as LangChain, AutoGen, or the Claude Agent SDK) and make principled tool-selection decisions that the broader engineering organisation can build on.
• Own AI cost efficiency across the agentic squad, monitoring token usage, model selection, caching strategies, and prompt optimisation to ensure RightShip gets maximum value from its AI investment without unnecessary spend.
• Coach the engineering team on agentic patterns, prompt engineering, and when not to use AI, building capability and confidence across the organisation as adoption grows.

Qualifications, Skills & Experience

• Master's or Bachelor's degree in Computer Science, Software Engineering, Artificial Intelligence, or an equivalent discipline.
• 10+ years of experience as a software engineer, with at least 2 years working directly with LLMs or AI-powered systems in a production context.
• Proven ability to distinguish deterministic from non-deterministic workflow steps, knowing when to use structured code versus an AI model, and to explain those decisions clearly to both engineers and non-technical stakeholders.
• Hands-on experience building and operating agentic or multi-step AI pipelines in production, using frameworks such as LangChain, AutoGen, CrewAI, the Claude Agent SDK, or comparable tooling.
• Strong proficiency in Python for AI and product development.
• Deep practical knowledge of prompt engineering, tool/function calling, retrieval-augmented generation (RAG), and structured output patterns.
• Experience integrating AI-powered automation into SDLC workflows, including code review bots, test generation, PR triage, CI/CD triggers, and similar.
• Familiarity with orchestration and evaluation tooling; able to measure, monitor, and improve agent output quality over time.
• Experience with cloud platforms, preferably Azure; able to reason about inference cost, latency, and reliability trade-offs.
• Familiarity with vector databases (e.g. Azure AI Search, Pinecone, Weaviate) and embedding pipelines.
• Strong analytical and communication skills; able to operate with autonomy in a fast-moving, globally distributed team and bring clarity to ambiguous problems.
• Genuine passion for building systems that reduce toil, extend engineering capacity, and keep humans meaningfully in control.

Tech Stack

The Systems Agentic Engineer operates across the engineering organisation and is expected to be largely stack-agnostic. The value is in the agentic and AI expertise, not deep knowledge of any specific language or framework. The following is provided as context so candidates understand the environment they will be working within: