AI Principal Engineer

About Brightpath Technologies Brightpath Technologies works with top enterprises and digital native businesses in Australia helping them build their remote teams in India, Philippines and Sri Lanka. We pride ourselves with hiring great teams to work on exciting game changing technology. Brightpath Technologies currently has 200+ engineers, testers, product managers etc. working across 20+ partners. We focus on uplifting the way organisations build and manage their remote teams through great working relationships, trust, integrity, culture and hiring processes. Additional to this we are building our own technology product offerings. We strive to deliver the best outcomes for our customers, our partners and our people. About our Client: Our client builds secure infrastructure for business payments — helping companies pay suppliers by card or bank transfer, regardless of how the supplier accepts payment. We reduce friction in B2B payments and improve working capital for businesses moving real money through regulated channels. They partner with payment networks, banks, and enterprise customers. Reliability, security, auditability, and disciplined engineering matter as much as delivery speed. Their platform spans a mixed technology estate — legacy and modern components across frontend, backend, integrations, and infrastructure — which is exactly why we need a thoughtful approach to AI-assisted development, not a blanket mandate to "use Copilot everywhere." About the job: We are hiring a hands-on Principal Engineer to lead our new AI-first engineering pod — a small, high-trust team that ships real product and platform work in our existing payments codebase, then helps the rest of engineering adopt what works. This is not an advisory architecture role. You will code, review, unblock, and ship — while building the AI-first operating model from evidence, not slides. Framework work should emerge from what the pod proves in production: tooling choices, review practices, guardrails, and team habits that other squads can pick up without reinventing the wheel. Over time, as the model proves itself, the pod is expected to grow into a full scrum team and potentially split into two scrum teams, as delivery capacity and adoption mature. You should be comfortable starting lean and hands-on, then evolving into a team lead who can run backlog, ceremonies, and engineering practices at squad scale. AI-frst means agent-first development with human checkpoints where they matter— orchestrated pipelines, not a developer alone in a chat window. Work flows through defined steps: intake → context assembly → agent execution → automated review gates → human approval → production. Using Cursor or Copilot well is a baseline. We want someone who has designed and built agentic orchestration: pipelines that connect real systems (Slack, issue trackers, Git, CI/CD) and run multi-step agent workflows with guardrails, audit trails, and human escalation at the right points. It does not mean unmanned codegen or bypassing regulated change control. How the team works: You will start with a lean pod (you plus a handful of engineers) owning a bounded slice of product or platform delivery. The pod is a reference implementation, not a silo. Success includes what the wider organisation learns from it. | Phase | Shape | | 0–6 months | Lean pod — prove the workflow, draft the playbook | | 6–12 months | Full scrum team — sustainable velocity and rituals | | 12+ months | One or two scrum teams — delegation and org-wide champions | You will bring existing engineers along through pairing, short secondments, open working sessions, and a living playbook on tools, guardrails, and review practices. Expect roughly half your time on pod delivery, a quarter on codifying and teaching, and the remainder on org-wide influence — tool approval, guardrails, and high-risk design review. Key Responsibilities: Lead pod delivery end-to-end: backlog refinement, technical breakdown, implementation, review, and release. Design and build agentic orchestration pipelines — e.g. a bug reported in Slack or Plane flowing through triage, context gathering, fix attempt, PR creation, and automated review before a human merges. Wire agentic review triggers into PR and CI/CD workflows: security analysis, bug-risk review, test gap detection, dependency checks — with clear pass/fail/escalate behaviour and auditability. Use AI-assisted development responsibly across coding, testing, debugging, refactoring, documentation, and code review — coaching the pod without bypassing engineering fundamentals. Ship production-visible outcomes early and codify what works into standards, guardrails, and tooling choices the wider org can adopt. Provide pragmatic technical leadership on pod-owned work and high-risk cross-cutting decisions; advise (not own) wider platform architecture. Lead one high-leverage modernisation path the pod can execute — including assessment of our .NET Core 2.1 estate and a pragmatic upgrade recommendation. Improve the pod’s