About Codex
Codex is an Australian technology consultancy delivering high-quality data, cloud, and AI solutions for organisations across sectors including healthcare, energy, and financial services.
We are a high-performing, fast-paced consultancy that values people who bring energy, accountability, and pride in their work. We care deeply about delivery — not just getting things done, but getting them done properly and in ways that genuinely help our clients succeed.
We are also officially Great Place to Work® Certified, reflecting the strong culture we’ve built — one grounded in trust, high standards, ownership, and genuine support for our people.
If you enjoy solving complex technical problems, building intelligent systems from the ground up, and working at the frontier of applied AI, you’ll fit in well here.
The Role
We’re looking for a Senior AI Engineer to design and deliver production-grade, agentic AI systems for enterprise clients.
You might come from:
- A strong software engineering background now working with AI systems, or
- ML/AI engineering background with strong experience in deploying models into production.
This role is about building reliable, secure, real-world AI systems — not just experimentation. You’ll design and implement autonomous and semi-autonomous AI systems that reason, orchestrate tools, interact with enterprise data platforms, and operate in secure cloud environments.
What You’ll Do
Build Agentic AI Applications
- Design and implement LLM-powered applications and AI agents
- Build multi-step reasoning workflows with tool use, memory and MCP (Model Context Protocol)
- Develop advanced RAG architectures integrating AI models with enterprise data
- Engineer robust backend systems that support AI-driven workflows
Production-Grade Engineering
- Deploy AI systems into production environments on AWS with evaluation, logging, monitoring, and guardrails
- Optimise for reliability, performance, and cost
- Apply strong software engineering discipline — testing of prompts/agents, integration tests, validation pipelines, and CI/CD
- Strong focus on compliance in production (e.g. content filters, PII redaction, Bedrock Guardrails or equivalent)
Cloud & Architecture
- Architect scalable solutions using AWS services such as Bedrock, SageMaker, Lambda, ECS/EKS, Step Functions, and API Gateway
- Implement secure IAM, networking, and data access patterns
- Contribute to reusable AI and cloud patterns across engagements
Client Delivery
- Translate business problems into AI-native solution designs
- Act as a trusted technical partner
- Work closely with clients and cross-functional delivery teams
- Communicate clearly and deliver with accountability
What We’re Looking For
Strong Engineering Foundations
- 7–10+ years of experience in software engineering, AI engineering, or related fields
- Strong Python (or equivalent backend language) experience
- Experience building and deploying production systems
- Solid understanding of APIs, distributed systems, and modern backend architecture
- CI/CD: GitHub Actions or Bitbucket Pipelines; Infrastructure as Code at scale (AWS CDK and/or Terraform)
- Testing: unit and integration tests; experience testing non-deterministic AI behaviour (prompts, retrieval, agent flows)
AI Exposure (Practical, Not Just Theoretical)
- Experience building LLM-powered applications, RAG systems, or AI-driven features or
- Strong interest in applied AI with demonstrable hands-on projects
- Understanding of embeddings, vector databases, and model evaluation is highly regarded
Cloud Experience
- Experience delivering production workloads on AWS
- Familiarity with services such as Lambda, VPC, IAM, S3, Bedrock, or SageMaker
- Infrastructure as Code experience (Terraform or similar) is beneficial
GenAI & RAG
- GenAI platforms: Amazon Bedrock, OpenAI, Anthropic Claude — model selection and trade-offs
- Vector stores: OpenSearch, Amazon OpenSearch Serverless, pgvector, or Pinecone for RAG and semantic retrieval
- Advanced RAG: ingestion, chunking strategies (semantic, hierarchical, hybrid), embedding model selection, retrieval tuning, evaluation (recall, precision, latency)
- RAG/agent frameworks in production
Agentic AI
- Tool use, orchestration, planning/execution patterns, disciplined justification for when to use agents vs simpler flows
- MCP: production-grade tool integration and context management; clear boundaries (tool contracts, error handling, fallbacks)
Observability, guardrails, compliance & security
- GenAI observability: Langfuse and/or MLflow for tracing, evaluation runs, and lifecycle management
- Guardrails in production: content filters, PII redaction, policy checks (e.g. Bedrock Guardrails)
- Compliance in production: define and apply data residency, audit logging, retention, and alignment with regulatory or client policies (e.g. sector requirements); awareness of frameworks or obligations relevant to engagement context
- Secure patterns: IAM least privilege, Secrets Manager/Parameter Store, network isolation
Ways of Working
- High energy and strong ownership mindset
- Deep care for quality and delivery excellence
- Comfortable with ambiguity in emerging AI domains
- Confident in client-facing, fast-paced environments
- First-principles thinking when evaluating GenAI, agentic systems, and architectural trade-offs
Why Codex
Great Place to Work® Certified – A culture built on trust, ownership, and high standards
High-Performance Environment – Meaningful responsibility and real impact
Frontier AI Work – Build production-grade agentic systems, not just prototypes
Strong Team Culture – Smart, supportive people who care about doing great work
Growth Opportunity – Accelerate your AI capability within a consultancy environment
Flexibility – Hybrid working in Melbourne
Pay: $140,000.00 – $200,000.00 per year
Work Location: Hybrid remote in Melbourne VIC
