Senior Backend AI Engineer

AutoGrab is on a mission to reshape the automotive industry with data-driven, intelligent technology. Backed by leading investors and trusted by major OEMs, marketplaces and dealer groups, we're building the infrastructure powering the future of vehicle trading.We have been recognised in the AFR Fast Starters, LinkedIn's top 25 Startups and Deloitte Tech Fast 50 lists, reflecting our rapid growth, strong market demand and relentless pursuit of innovation.About the RoleThis is a senior engineering role at the intersection of backend engineering and agentic AI. You will work within AutoGrab's engineering team, reporting to the Tech Lead - Backend Engineering, to build and operate LLM-driven backend applications and agentic workflows that sit at the core of our product intelligence.We are looking for someone who is genuinely cross-disciplinary, equally at home in the backend service context as in the agentic LLM space. You will bring strong applied experience in building production agentic systems, combined with deep backend engineering fundamentals across deployed services and data-intensive applications.This is a hands-on engineering role. You will write production code, shape technical decisions and contribute meaningfully to a small, high output team that moves quickly and values quality. While you will also use agentic tooling to write code, you'll need to be able to understand and write code by hand, as well as craft the details of an architecture and technical solution design.Key ResponsibilitiesBackend EngineeringWrite, review and understand well-structured, tested, production-quality code, stepping in with your own judgement where agentic coding tools fall shortBuild and maintain production-grade backend services that serve and process AutoGrab's live dataWork across a variety of databases, SQL and data-intensive applications, both stream and batch orientedAgentic AI EngineeringDesign, build and operate agentic LLM workflows and data applications in productionApply prompt engineering, RAG and vector search, and tool/function calling in a data pipeline or data analysis contextOwn LLM evaluation and observability, and manage inference scalably, reliably and cost-effectivelyContribute to guardrail design and compliance-sensitive flow handling within AutoGrab's AI systemsInfrastructure & OperationsDeploy and operate LLM-related services on GCP, including Vertex AI, Cloud SQL and Cloud Run/GKEWork with infrastructure-as-code (Terraform) and CI/CD pipelines to maintain operational disciplineContribute to system reliability, cost management and observability across production servicesCollaboration & CommunicationWork through ambiguity with the team and technical leadership, making trade-offs clearly and setting expectations honestlyCollaborate closely with engineers across backend, data science and product disciplinesContribute to engineering discipline: code review, testing standards and technical documentationAbout YouRequiredHands-on programming and solution design experience, and genuine care for the engineering craftExperience building and deploying production-quality backend servicesStrong experience with a variety of databases, SQL and data-intensive applications, both stream and batch orientedGenuinely cross-disciplinary across agentic LLM-driven and backend applications, this is the core of the roleStrong applied experience building agentic LLM workflow systems in production: prompt engineering, RAG, vector search and tool/function calling, ideally in a data pipeline or data analysis contextSolid understanding of LLM evaluation, observability and how to manage inference scalably, reliably and cost-effectivelyExperience deploying and operating LLM-related services on GCP (Vertex AI, Cloud SQL, Cloud Run/GKE), with infrastructure-as-code (Terraform) and CI/CDExcellent communication and collaboration skills, comfortable working through ambiguity, making technical trade-offs and setting expectations honestly with the team and leadershipPreferredExperience with LLM fine-tuning or training embedding modelsExperience using LLMs for search, and evaluating search systemsFamiliarity with evaluation-driven development for LLM products and guardrail/safety patterns for compliance-sensitive use casesFamiliarity with Helm, ArgoCD or KubernetesExperience with Python data science and data analysis toolsWhy AutoGrabJoin a high-growth technology company building something genuinely novel at the intersection of automotive data and AIWork in a small, high-output engineering team where your contribution has direct product impactHands-on, collaborative culture with real exposure to agentic AI systems in productionReport directly to the Tech Lead -Backend Engineering, with clear lines to senior technical leadershipOffices in Australia, the UK and Asia with global expansion under wayCompetitive remuneration and the opportunity to grow with the business