You'll take LLM applications and agents past the prototype and into production, in an environment where they have to be trustworthy.
The role
- Build LLM-powered applications and agentic workflows in Python on Amazon Bedrock and AgentCore, using foundation models including Claude and frameworks such as Strands Agents or LangGraph
- Engineer retrieval and context: ingestion, chunking, embeddings, hybrid search, and MCP servers connecting models to enterprise tools and data
- Make quality measurable with evaluation datasets, automated evals and regression tests for prompts, retrieval and agent behaviour
- Run AI like real software with IaC, CI/CD, tracing and cost monitoring, plus guardrails, PII handling and audit logging
- At Lead level and above: own AI platform architecture, take use cases such as customer servicing, lending and financial crime to production, and grow a team
- Senior AI Engineer: 3-6 years, including 1-2+ years building ML or GenAI. Lead: 6-10 years, including 3+ in ML/GenAI. Senior Lead: 10-15 years, typically with 4+ in ML/GenAI
- Strong Python and software engineering fundamentals
- Hands-on LLM work on AWS (Bedrock or SageMaker AI), covering RAG, prompt design and tool use
- At least one AI solution you've taken past proof of concept into production and kept running
- Financial services or regulated industry experience is preferred
Culture
- An AI engineering community at the front of what these platforms can do, sharing what works
- Mentoring and coaching at every level
- Hybrid in Sydney, with time on site with the team at agreed intervals
- $120k-$240k package, depending on level
- Permanent, or contract at $800-$1,200 per day
- Support to gain AWS AI and ML certifications
- A fast process: CVs reviewed within 48 hours and no take-home or live coding test
