Forward Deployed Engineer

The Unmarked Group and our AI division AutoSolutions.ai builds AI-powered software for real business operations: internal platforms, AI agents, workflow automation, lead-generation systems, client portals and reusable white-label products. We work from inside the businesses themselves, sitting with operators, understanding how the work actually happens and removing the manual parts.

We are a small, hands-on team. Projects move quickly and you are in the room while they do, but the software still needs to work after the demo. The strongest patterns discovered in one deployment are fed back into our core products and reusable blueprints.

Forward deployed means going where the problem is. You will embed with one business at a time and shape the solution alongside the people doing the work. This is a customer-facing engineering role at the point where our products meet real operations.

The work is closing the implementation gap between our general platform and each business's legacy systems, operating rules and existing tools, then feeding what you learn back to the product team. You need enough experience and judgment to hold a deployment on your own while escalating the decisions that need senior input.

This is a full-time, on-site role. You will work from our Melbourne office and, when a deployment needs it, from the businesses we build for. It is not remote or hybrid.

• Embedded Discovery & Problem Definition

  • Embed with a business, sit with its operators and map how the work actually happens today.

  • Identify the operational constraint, scope the smallest useful solution and agree what successful adoption looks like.

  • Be the technical person in the room with operators and business owners, translating between real workflows and product decisions.

• Deployment & Integration

  • Shape solutions across frontend, backend, APIs, databases and data flows.

  • Fit our platform to legacy systems, compliance rules, rigid approval processes and existing tools.

  • Deploy integrations into CRMs, ERPs, SaaS tools, legacy databases and third-party data sources.

  • Build full-stack applications for the teams sitting next to you, including lead-generation, enrichment and reporting workflows.

• AI Agents & Automation

  • Deploy AI agents that use a business's own tools, documents and data.

  • Develop AI workflows using tool calling, structured outputs, retrieval, memory and automation.

  • Use secure browser automation where APIs fall short.

  • Turn successful one-off customisations into repeatable blueprints that can improve the core platform.

• Reliability, Adoption & Product Feedback

  • Hold the line on authentication, permissions, audit trails, monitoring, error handling and data integrity.

  • Harden prototypes until they are safe and reliable enough for daily production use.

  • Demo on site, train the people using the solution and gather feedback directly from them.

  • Document the deployment clearly and feed recurring needs, patterns and product opportunities back to the core team.

• Qualifications & Competencies:

  • A problem-solver's mindset and a habit of investigating until things work.

  • Product sense: the ability to generalise a one-off workaround into a repeatable tool or blueprint.

  • Genuine obsession with AI, experimentation and what can now be built with it.

  • Deep understanding of frontend and backend systems, APIs, databases, authentication and integrations.

  • Comfort being the technical person in the room with operators and business owners.

  • Confidence using AI coding tools to build, inspect, test and improve software.

  • Consultative judgment: the ability to read constraints, make sound trade-offs and explain them clearly.

  • Clear communication with technical and non-technical people, especially in person and on the spot.

  • Australian working rights, based in Melbourne and willing to work on site with the businesses.

We expect you to use Codex, Claude Code, Cursor and similar tools. Hand-coding is not the test; understanding the system, solving the problem, reviewing the output and taking responsibility for what ships are. You do not need to be able to write code unaided, but you must understand software-development best practices and be able to evaluate what the tools produce.

• Agent systems: tool calling, MCP, agent skills, retrieval, memory or orchestration.

• Product stack: Supabase, PostgreSQL, Vercel, Cloudflare, AWS or similar platforms.

• Automation: Playwright, browser automation, n8n, Make, Zapier or Pipedream.

• Legacy environments: old databases, on-premise tools, rigid approval processes and compliance rules.

• Multi-tenant SaaS, white-label products, Electron, data enrichment, role-based access control or audit logs.

• First 30 Days

  • Understand our products, the businesses we serve, our workflow and our technical stack.

  • Set up the main projects and ship a useful production improvement.

  • Sit in on operator conversations and learn how we translate business problems into software.

• By 60 Days

  • Own one deployment end to end, from the operators' problem through to adoption.

  • Gather feedback inside the business and fix the operational problems you find.

  • Use AI development tools without sacrificing judgment or quality.

• By 90 Days

  • Take ownership of a business's product surface or a substantial workflow.

  • Ship with appropriate tests, documentation and monitoring.

  • Work independently while escalating the decisions that need senior input.

• Work directly with the founder and influence what we build.

• Own real deployments rather than isolated tickets.

• Take what you learn on site back into the core platform.

• Watch your work change how a team operates, from inside the room.

• Grow toward a lead forward-deployed or AI solutions role as the team expands.

Competitive base salary plus superannuation, benchmarked as an engineering role rather than a delivery role. This is a full-time, on-site position based from our Melbourne office and, when required, the businesses we build for. It is not a remote or hybrid role.

Send apply now including you CV and LinkedIn profile, links to work you can discuss, and a short description of one business process you turned into software.

1. What was the original problem?

2. What did you build and which technology did you use?

3. What went wrong, and what changed once people used it?

1. AI screening with one of our HR AI agent interviewers.

2. Meeting with the Head of AI covering your background, systems thinking and product judgment.

3. In-depth final conversation covering the work, expectations and growth.

AutoSolutions.ai welcomes applicants from different backgrounds and career paths. We assess candidates on practical ability, judgment, communication and the quality of their work. A short, specific application is more useful to us than a generic cover letter.