Software Engineer – AI Agents & Agentic Systems

About the Role

We are looking for a hands-on Software Engineer who is passionate about building with AI. You will work at the intersection of LLMs, AI coding harnesses, autonomous agents, software engineering, and cloud infrastructure, helping us build both our core product and deliver technical solutions for clients.

We are particularly interested in engineers who are already using tools such as Claude Code, OpenAI Codex, and other AI-powered development environments to build, test, debug, and ship real software. You understand that the opportunity extends well beyond using AI to generate snippets of code: it is about designing effective agentic workflows, orchestrating multiple tools and models, automating complex tasks, and building reliable systems that can operate with increasing levels of autonomy.

You will help us explore and implement new ways of engineering software using LLMs and agent harnesses, while developing the underlying platforms, integrations, and infrastructure needed to make these systems secure, observable, and production-ready.

Our work spans AI-enabled applications, agent orchestration, developer tooling, cloud-native platforms, and infrastructure software. This includes building systems that allow organisations to understand, observe, govern, and audit workflows executed by AI agents — including the tools they invoke, the decisions they make, the data they access, and the actions they take.

Alongside product development, you will contribute to client engagements involving AI engineering, software development, automation, and infrastructure. These projects will give you opportunities to apply emerging AI techniques to real-world engineering problems across different environments and technology stacks.

Our primary technology stack includes Go, Python, TypeScript, AWS, and Kubernetes, complemented by a growing ecosystem of LLMs, AI coding harnesses, agent frameworks, APIs, and developer tools.

We are a small, focused engineering team, so this is a role for someone who enjoys building, experimenting, learning quickly, and taking ownership from concept through to production. We value engineers who can work independently, make sound technical decisions, and combine the speed of AI-assisted development with the rigour required to build reliable software.

Key Responsibilities

1. AI-First Software Engineering & Coding Harnesses

• Use AI coding harnesses such as Claude Code and OpenAI Codex as core tools in your daily engineering workflow.

• Design and refine workflows in which AI agents can explore codebases, plan implementations, modify code, run tests, debug failures, and complete multi-step engineering tasks.

• Work with agent instructions, project context, skills, tools, reusable workflows, and other harness capabilities to make AI-driven development more effective and repeatable.

• Build and maintain software using an AI-first approach, applying engineering judgement to review, validate, test, and improve AI-generated code.

• Experiment with different models, harnesses, configurations, and approaches to identify where AI agents can meaningfully improve engineering speed and quality.

• Develop repeatable workflows for specification, implementation, code review, automated testing, debugging, and deployment.

• Establish appropriate guardrails around permissions, secrets, code changes, tool access, and execution environments.

• Help evolve our engineering practices as AI coding tools and agent capabilities develop.

2. AI Agents, LLMs & Agentic Workflows

• Design, build, and integrate AI agents that use LLMs to reason through tasks, interact with tools, retrieve information, and execute multi-step workflows.

• Develop applications powered by commercial and open-source LLMs, using model APIs and appropriate orchestration patterns.

• Build tool-using agents that can interact with APIs, databases, code repositories, cloud services, and external systems.

• Experiment with agent harnesses, context management, tool calling, structured outputs, memory, planning, and multi-agent coordination where appropriate.

• Develop effective approaches to retrieval-augmented generation (RAG), embeddings, semantic search, and knowledge retrieval when required by the use case.

• Evaluate model performance, latency, cost, reliability, and task completion, choosing appropriate models and architectures for different requirements.

• Design workflows that account for the non-deterministic nature of LLMs, including failure handling, retries, validation, human approval, and controlled execution.

• Build evaluation and testing approaches to measure agent performance and prevent regressions as prompts, models, tools, and workflows evolve.

• Translate promising AI prototypes into maintainable, secure, and observable production systems.

3. Product Engineering & AI Infrastructure

• Design, build, test, validate, deploy, and maintain our core product and supporting services across the full software development lifecycle.

• Develop cloud-native applications, backend services, APIs, integrations, and infrastructure components.

• Contribute to platforms that provide visibility, governance, security, and auditability for AI-powered systems and agentic workflows.

• Build mechanisms for tracking agent activity, tool invocations, execution history, model interactions, permissions, and outcomes.

• Help organisations understand what their agents are doing, which systems they interact with, and how their actions can be traced and reviewed.

• Contribute to identity, authentication, authorisation, policy enforcement, and data lineage capabilities for agentic systems.

• Make pragmatic architectural decisions that balance delivery speed with reliability, security, scalability, and maintainability.

4. AI-Enabled Software Development Lifecycle

• Explore how AI agents can support and automate activities across the entire software development lifecycle.

• Develop workflows that connect requirements, planning, implementation, testing, code review, deployment, and maintenance.

• Build systems that enable agents to work with existing repositories, development tools, CI/CD pipelines, and engineering platforms.

• Implement automated validation, regression testing, and verification to ensure AI-generated changes meet defined requirements.

• Investigate opportunities for agents to diagnose defects, investigate production issues, analyse logs, and propose or implement fixes within controlled boundaries.

• Help establish practical approaches for human oversight, approval gates, reproducibility, and accountability in AI-assisted engineering.

• Share learnings and develop reusable patterns that make AI-driven engineering workflows more reliable and effective.

5. Cloud Infrastructure & Platform Engineering

• Build and deploy AI-enabled applications and services across AWS and other cloud or self-hosted environments.

• Develop the infrastructure required to run agents, model integrations, background jobs, and long-running workflows.

• Work with Docker, Kubernetes, CI/CD pipelines, infrastructure as code, and deployment automation.

• Build secure execution environments and sandboxing approaches for agents that run code, invoke tools, or interact with external systems.

• Contribute to observability, logging, monitoring, tracing, and cost management across applications and agentic workflows.

• Troubleshoot issues across application, model integration, networking, infrastructure, and distributed system layers.

• Help ensure AI-powered systems can operate reliably, securely, and economically in production.

6. Client Consulting & Technical Delivery

• Deliver software engineering and AI implementation projects for clients across a variety of technical environments.

• Help clients identify practical opportunities to use LLMs, AI coding harnesses, and autonomous agents to improve software development and business workflows.

• Design and build AI-powered applications, agentic automations, integrations, and developer tooling.

• Evaluate existing systems and identify opportunities to introduce AI capabilities safely and effectively.

• Prototype, validate, and deploy solutions that address real business and engineering challenges.

• Adapt to different client requirements, architectures, and technology stacks while maintaining high standards of engineering quality.

• Communicate technical trade-offs, implementation approaches, and AI-related risks clearly to technical and non-technical stakeholders.

• Take ownership of deliverables and help clients move beyond proof-of-concept demonstrations to useful, production-ready solutions.

7. Collaboration, Experimentation & Open Source

• Work closely with a small team of engineers and builders, contributing across AI engineering, product development, and infrastructure.

• Stay current with developments in LLMs, agent harnesses, AI development tools, and emerging agent architectures.

• Experiment with new approaches, share findings, and turn successful experiments into reusable engineering practices.

• Participate in architectural discussions, code reviews, technical design, and continuous improvement.

• Where relevant, contribute to open-source initiatives, including Linux Foundation projects and wider developer communities.

• Challenge assumptions constructively and help foster a culture of technical curiosity, accountability, and continuous learning.

Technical Skills & Experience

We are looking for an engineer with strong software development fundamentals who is excited about the shift towards AI-first engineering. You do not need to be an expert in every technology listed below, but you should have practical experience in several areas and a willingness to learn quickly.

AI Engineering & Agentic Development

We are particularly interested in candidates who have hands-on experience with some of the following:

• AI coding harnesses: Claude Code, OpenAI Codex, or comparable agentic coding environments.

• Agent development: Building agents that can use tools, interact with APIs, execute tasks, and work through multi-step problems.

• LLM integration: Working with model APIs, including OpenAI, Anthropic, or other commercial and open-source providers.

• Agent architecture: Experience with tool calling, context management, structured outputs, agent state, orchestration, and human-in-the-loop workflows.

• AI development workflows: Using agents for code generation, refactoring, testing, debugging, repository analysis, and other engineering tasks.

• Evaluation and reliability: Testing LLM outputs, evaluating agent performance, handling failures, and building guardrails around model behaviour.

• Retrieval and knowledge systems: Familiarity with embeddings, vector search, RAG, or other techniques for grounding LLM responses in relevant information.

• AI frameworks and tooling: Experience with agent frameworks, orchestration libraries, model gateways, or MCP-based tool integrations is valuable.

We are interested in practical capability rather than familiarity with a long list of frameworks. The ability to build a useful agent, understand its limitations, and make it reliable is more important than having used a particular library.

Software Engineering & Core Technologies

• Go (Golang): Experience building backend services, APIs, infrastructure tooling, or distributed systems.

• Python: Experience developing applications, automation, integrations, agent workflows, or AI-related tooling.

• TypeScript: Experience building applications, services, APIs, or user-facing components.

• AWS: Familiarity with cloud services, deployment, infrastructure, and cloud architecture.

• Kubernetes and Docker: Experience deploying, operating, or supporting containerised applications.

We do not expect every candidate to be equally strong across the entire stack. Depth in one or more languages, combined with solid engineering fundamentals and the ability to work across technologies, is more important than ticking every box.

Additional Valuable Experience

• Building production software in startups, small engineering teams, or product-focused environments.

• Developing cloud-native platforms, distributed systems, infrastructure software, or developer tooling.

• Building autonomous workflows or integrating LLMs into existing applications and enterprise systems.

• Working with CI/CD, infrastructure as code, automated testing, observability, and production monitoring.

• Designing secure systems involving identity, permissions, audit trails, and data governance.

• Developing sandboxed execution environments or systems that allow software agents to interact safely with infrastructure.

• Working with REST APIs, databases, message queues, and third-party integrations.

• Contributing to open-source projects or collaborative technical communities.

What We're Looking For

Above all, we want a builder who is excited about what becomes possible when strong software engineering is combined with capable AI agents.

You are likely to be a good fit if you:

• Actively use tools such as Claude Code or Codex to build real software, rather than treating AI as an occasional coding assistant.

• Understand how to structure tasks, provide context, guide agents, inspect their work, and verify their outputs.

• Enjoy building AI agents and LLM-powered applications that do useful work, not just experimenting with chat interfaces.

• Have strong programming fundamentals and can recognise when AI-generated code is incorrect, insecure, inefficient, or unnecessarily complex.

• Are comfortable working across the stack, from application code and APIs through to cloud infrastructure and deployment.

• Take ownership of problems and can move from an ambiguous idea to a working, tested, production-ready solution.

• Are curious, pragmatic, and willing to experiment with new models, tools, and engineering approaches.

• Understand that production AI systems require evaluation, observability, security, and clear boundaries around autonomous execution.

• Enjoy working in a small team where initiative, adaptability, and technical judgement matter.

• Can communicate clearly with colleagues, clients, and stakeholders, and explain complex technical decisions in practical terms.

Experience in consulting, professional services, startups, or product engineering would be beneficial.

What Makes This Role Interesting

This is an opportunity to work on the practical application of AI to software engineering and the development of agentic systems.

You will have the opportunity to:

• Build with AI at the centre: Use modern AI coding harnesses and LLMs as fundamental parts of your engineering workflow.

• Create agents that do real work: Design and build systems that interact with tools, execute workflows, and solve meaningful problems.

• Work on the next generation of software engineering: Explore how agentic development changes the way software is designed, built, tested, and maintained.

• Build the infrastructure behind AI: Contribute to platforms that make agentic systems observable, governable, secure, and suitable for production.

• Work across product and consulting: Apply your skills to our own platform and a variety of real-world client challenges.

• Take meaningful ownership: Work in a small team where you can influence architecture, tooling, engineering practices, and product direction.

We are looking for engineers who want to help shape how software is built in an AI-first world — combining the speed and leverage of LLMs with the discipline, creativity, and judgement of excellent software engineering.