Senior Backend Engineer - AI Platform

EnterpriseAI is the platform consultants use to build enterprise AI systems for their clients, and to put them into production inside the client's own systems. Who we work with Software vendors adding AI workflows into their own product, under their own brand. Specialist consultants delivering AI process-improvement solutions to their clients. Enterprises buying the platform directly to build and run their own AI workflows. The problem A consultant builds a client app in Claude or ChatGPT over a weekend. On Monday the client's leadership loves it. Six weeks later it still isn't signed off, because the client's IT team asked three questions nobody could answer: Who can see this app, and the data in it? What happens when the business changes, or the AI model does? Who maintains it, and how do our people learn to? Our platform answers them. It runs inside the client's own Azure tenant behind their own logins, so the data never leaves. The workflow and the model can be changed without rebuilding the app or touching the systems underneath. The consultant who built it maintains it on the same platform. It is in production today for a NSW local government service, a software vendor is taking it to its customers under its own brand, and consultants are selling it to their enterprise clients. You would build the backend that makes those three answers hold as the number of consultants, clients and tenants grows. What you'd build in the next six months We are launching to consultants in October. Each one builds an app for one of their own clients, and some will take it into production. The backend work behind that: The CLI and its installers on Mac, Windows and Linux, and the deploy path from Claude Code and Codex. The services behind the no-code builder: AI connections, document understanding, content extraction. Multi-tenant hierarchy, identity, permissions and audit logging that hold up in enterprise and government procurement. Model-agnostic routing, evaluation and monitoring, so a model can be swapped without the app noticing. Curated per-process context with write-back to the client's systems of record, and connectors to the systems consultants' clients actually run. Stripe, subscription plans and usage metering. The reliability and observability work that lets a wave of new users arrive in one month. You would own features end to end, from technical design to production, and you would set the standard for the engineers around you. What you'd own Backend services, APIs, integrations and data flows on the platform. Features from design through production, including the operating of what you ship. Code review and practical technical direction for engineers in Sydney and Manila. Architecture decisions sized for a product that changes weekly, and the judgement to know which ones are reversible. Spotting technical risk before it becomes a delivery problem. What we're looking for Around 5 to 10 years of professional software engineering, most of it backend. Deep, current Python. FastAPI is a plus. You have designed, built and operated APIs, services and data models in production, on a cloud platform. Azure preferred; AWS or GCP relevant. You have integrated with enterprise systems of record and know why that is harder than it looks. You have taken meaningful technical ownership rather than only working from predefined tickets. You review code in a way that makes other engineers better. You want to stay hands-on rather than move into management. Useful, not required Shipping an LLM-powered capability into production: LLM APIs, agents, RAG, evaluation, guardrails, human-in-the-loop, AI observability. Building developer platforms, workflow engines, orchestration systems or multi-tenant SaaS. TypeScript and Next.js. MongoDB. A startup or small product team, where you saw your work reach users. Our stack Backend: Python and FastAPI. Frontend: TypeScript and Next.js. Cloud: Microsoft Azure, deployed into each client's own tenant. Data: MongoDB, with curated per-process context and write-back to the client's systems of record. Platform: multi-tenant hierarchy, identity and audit, workflow orchestration, model-agnostic routing, evaluation and monitoring, enterprise integrations, a CLI and the Configurator no-code builder. AI: LLM APIs from several providers, agents, RAG, document understanding. Who you'd work with The Head of Engineering, who you report to, and who stays hands-on in code review and design. The CTO above them owns architecture and technical direction. A small engineering team in Sydney and Manila. Manila works to Sydney hours. The ADAPTOVATE consultants who use the platform with clients every week, and whose problems become your backlog. A founding team of former BCG partners and operators who have built and sold software companies before. Location and ways of working Sydney CBD, in the office five days a week. We are launching in October, the product changes daily, and the engineers, the Head of Engineering and the CTO make decisions in the same room on the same day. Compensation Market salary and participation in the EnterpriseAI employee share option plan (ESOP). How to apply Apply with your CV through LinkedIn We are particularly interested in hearing about: A backend system or product you personally helped build and ship. What you owned. A difficult technical decision you made. How you improved the engineers or team around you. If you like building, taking ownership and turning difficult technical problems into production software, we'd like to hear from you.