Staff AI Deployment Engineer, Australia

Description

The company runs an AI-driven network that helps governments and businesses protect trade and strengthen supply resilience while goods keep moving.

The implementation team takes the company's platform from signed contract to live use inside customer businesses. The work covers integrations, fitting the platform into existing workflows, and making sure the insights change real decisions.

This is a delivery group with deep technical skill. You will work with customers from trade compliance analysts to enterprise IT, and with engineering, data science and product teams to turn field findings into what gets built next.

Responsibilities

• Build and deploy agentic AI workflows, including multi-step agents, tool and function calling, and automations, that run customer processes on live data
• Connect agents to the right data sources, APIs and systems of record, and set the tools, permissions and guardrails they work within
• Write and refine prompts, system instructions and evaluation criteria so agents are reliable enough for production
• Add orchestration and human-in-the-loop checkpoints so workflows stay safe, auditable and easy to correct
• Track accuracy, cost and latency on deployed processes and tune them as customer data and needs change
• Lead implementations end to end, from scoping and configuration through integration, validation, go-live and handover to steady-state support
• Configure the platform against real customer data and workflows, closing the gap between current product behaviour and what the customer needs
• Build API integrations, data handoffs, automations and internal tools, using AI-assisted development to move quickly
• Measure done as adoption: the customer's team uses the system in their own process without you present
• Act as the main technical contact during implementation, running working sessions, demos and deep dives with stakeholders at every level
• Draw out requirements from customers who cannot yet fully state them, and turn business problems into concrete technical scope
• Explain platform behaviour, data quality issues and trade-offs to non-technical audiences, and relay customer constraints accurately to the company
• Coordinate internal engineering, data science, product and customer success teams around one shared plan for each implementation
• Escalate blockers to the team that can solve them, make the request precise, and follow it through to resolution
• Pass patterns, recurring friction and product gaps from the field back to the product organisation
• Create reusable playbooks, templates and documentation so each implementation goes faster than the last
• Own the roadmap for your accounts, sequencing work by customer value and delivery risk and deciding what ships now versus later
• Write clear requirements, specifications and acceptance criteria that engineering and data science can build against
• Set measurable success criteria for each implementation and track progress against them

Requirements

• At least 8 years in customer-facing technical delivery, such as implementation, engineering, professional services, technical program management or product management with hands-on scope
• A proven record of shipping complex technical projects with external stakeholders against real deadlines and real consequences
• Strong communication and coordination: you can run a customer session in the morning and align three internal teams in the afternoon
• Product instincts: you ask what problem is being solved and what the smallest viable solution looks like
• Comfortable moving forward with incomplete information and unclear scope
• Prefer automation and repeatable processes over manual heroics, and would rather fix a process than repeat it by hand
• Working proficiency in Python or a similar language, enough to read a codebase, write and debug scripts, build integrations and ship small features. You do not need to be a systems engineer.
• Fluent in SQL and able to explore and validate data to answer your own questions
• Regular use of AI-assisted coding tools such as Claude Code, Cursor or Copilot. You use them on unfamiliar code, prototype integrations and automate repetitive delivery work, and you check what they produce.
• Hands-on experience building or deploying agentic AI systems, such as LLM-based agents, tool and function calling, retrieval-augmented generation or multi-agent orchestration
• Skill in prompt engineering and agent evaluation: designing prompts and evals, diagnosing agent failures and iterating toward reliable production behaviour
• Familiarity with agent frameworks and protocols such as MCP or LangChain, and with connecting agents to external tools, APIs and data sources
• Familiarity with REST and Swagger/OpenAPI APIs, Git-based workflows and cloud-hosted systems
• Enough knowledge of data pipelines, data modelling and ML systems to scope work, diagnose issues and discuss designs with the engineers who build them
• Experience with global trade, supply chain, customs or trade compliance
• Background in enterprise software implementation at scale, across multiple stakeholders and multi-quarter programs
• Exposure to Spark, Databricks or large-scale data processing
• Experience building internal tools or automation that a team adopted
• Formal product management experience, including roadmaps, specifications and stakeholder prioritisation
• Day to day tools and stack: Python, SQL and Spark; Claude Code and other AI development tools; LLM agents, tool and function calling, MCP, retrieval-augmented generation and agent orchestration frameworks; Docker, Git, Kubernetes, Swagger/OpenAPI, AWS and Azure; Databricks, Elasticsearch, Postgres and Redshift

Conditions

The role is based in Australia, and you should be open to travelling onsite to the Canberra hub

Job offers come from the company's own email domain, and the company never asks for payment during interviews or onboarding