Location
Job description
As a Full Stack AI Agentic Developer, you will lead the development of autonomous systems that do more than just generate text; they solve complex problems through tool-use and multi-step reasoning; 482 Visa sponsorship is available for the right candidate. You will be responsible for the end-to-end stack, from building robust agentic orchestration layers to creating the frontend interfaces that allow users to oversee and direct autonomous workflows.
📋 Core Responsibilities
✅ Selection Criteria
💻 The Ecosystem
📋 Core Responsibilities
- Agentic Orchestration: Design and implement autonomous agents using frameworks like LangGraph, CrewAI, or AutoGen.
- Reasoning & Planning: Develop sophisticated prompt chains and "ReAct" patterns to enable agents to plan, reflect, and self-correct.
- Tool-Use Integration: Build secure backend interfaces (Python/FastAPI) that allow AI agents to interact with external APIs, databases, and legacy systems.
- Full-Stack Interface: Develop React/Next.js frontends designed for "human-in-the-loop" monitoring and intervention in agentic processes.
- Evaluation & Debugging: Perform deep analysis of agent traces (using LangSmith or Arize Phoenix) to identify and resolve logic loops or tool-call failures.
- Platform Security: Implement strict guardrails and sandboxed environments to ensure autonomous agents operate within organisational safety boundaries.
✅ Selection Criteria
- Engineering Depth: 5+ years of Full Stack experience, with at least 18 months focused on Agentic AI or complex LLM orchestration.
- Framework Mastery: Expert-level proficiency in Python and modern agentic frameworks (LangGraph preferred).
- System Design: Strong understanding of event-driven architecture and asynchronous processing (Redis, RabbitMQ).
- Autonomy Mindset: Proven experience building systems that handle non-deterministic outputs and "self-healing" logic.
- Cloud Infrastructure: Experience deploying agentic workloads on AWS or Azure using containerised patterns (Docker/Kubernetes).
💻 The Ecosystem
- Core Languages: Python (FastAPI), TypeScript (React/Next.js).
- Agentic Stack: LangChain, LangGraph, CrewAI, AutoGen.
- Intelligence Layer: OpenAI (o1/GPT-4o), Claude 3.5 Sonnet, Llama 3.
- Data & State: PostgreSQL (pgvector), Pinecone, Redis.