Applied AI Engineer

About the Role

Join DX1 to develop robust AI agent systems tailored for enterprises in sectors such as financial services, healthcare, government, and telecommunications. This role transcends simple prompt engineering or linking large language models (LLMs) to vector databases. Instead, it focuses on meticulous design of harnesses, loop engineering, context engineering, and specification-based development, ensuring AI agents operate reliably in production environments under strict audit standards without risking AI-related incidents.

This position serves as a stepping stone within DX1's career framework, leading into the Forward-Deployed Engineer (FDE) program. Successful candidates who demonstrate strong delivery skills and complete the certification track may advance conditionally to FDE roles, embedding directly with clients and taking ownership of end-to-end delivery.

Key Responsibilities

• Architect and implement high-quality production AI systems encompassing agentic architectures, orchestrating multiple AI models, Retrieval-Augmented Generation (RAG) pipelines, tool usage, and integrations with Managed Control Plane (MCP) systems.
• Develop and maintain evaluation frameworks that enable measurement and defensibility of agent performance in production, including evaluation suites, regression testing, and observability through OpenTelemetry-based tracing and monitoring.
• Design the operational harness surrounding the AI agent, including tools, permission sets, state management, and protective guardrails, as well as the iterative loop processes (planning, acting, observing, correcting) rather than focusing solely on prompts.
• Translate clear, maintainable specifications into reliable systems, taking projects from prototyping through deployment phases, continuous integration and delivery (CI/CD), and operational procedures documentation.
• Implement AI-specific security safeguards such as validating inputs, controlling outputs, detecting Personally Identifiable Information (PII), maintaining thorough audit trails, and defending against prompt injection attacks and undue agent autonomy.
• Create and contribute reusable components such as agent templates, integration patterns, and compliance toolkits to DX1’s shared asset repository, accelerating future projects.

Required Qualifications and Experience

• Minimum of two years’ experience in production-grade software engineering with demonstrable systems shipped and maintained.
• Hands-on expertise in constructing solutions atop large language models, including RAG methodology, agent design, prompt and context engineering, and solution evaluation.
• Proficient programming skills in Python and/or TypeScript.
• Familiarity with cloud platforms, preferably AWS, alongside containerized deployment practices.
• Experience in applying OpenTelemetry instrumentation for production environments to gain observability into tracing, with emphasis on agent or LLM workflow metrics such as tool usages, reasoning steps, latency, and operational costs.
• Capability to execute tasks based on scoped technical specifications independently, and to articulate technical decisions to non-technical audiences clearly.

Certification and Career Advancement

DX1 supports your advancement by funding certifications from the outset, including AWS Certified AI Practitioner and Anthropic’s Claude Certified Architect (CCA) Foundations within six months. These certifications are prerequisites to join the Forward-Deployed Engineer program. Additionally, DX1 holds AWS partnership accreditation specializing in generative AI technologies.

Additional Valuable Skills

• Experience deploying AI agents on AWS Bedrock or comparable managed model platforms beyond simply using raw APIs.
• Prior exposure to consulting roles or direct client-facing delivery experience.
• Awareness of Australian data privacy laws and governance frameworks concerning AI technologies.