Work Across
Get started
All jobs
Save

Senior AI Engineer, Agents & Applications

Firmus
Contract type
Ongoing
Work mode
On-site · Sydney, Australia
Experience
Senior · 5+ years

Job description

Defines and implements the agent architecture and engineering framework for Firmus's AI infrastructure, covering orchestration, retrieval, tool use, evaluation, safety controls, observability, and deployment. The role builds specialist agents and RAG pipelines for Model-to-Grid and AI-factory domains.

At a glance

Employment
Full-time · Employee (employment agreement)
Work mode
On-site · Sydney, Australia

What you'll do

  • Define reference architectures for single-agent, multi-agent, workflow-based, event-driven, and human-in-the-loop agentic systems.
  • Select the appropriate architecture for each use case rather than applying multi-agent patterns by default.
  • Build specialist agents for Model-to-Grid and AI-factory domains, such as benchmark-analysis, workload recipe, inference-endpoint selection, Kubernetes and job-scheduler diagnostic, GPU-topology, AI-factory health, documentation, thermal, power, and grid-integration agents.
  • Build Retrieval-Augmented Generation (RAG) pipelines using vector retrieval, hybrid search, metadata filtering, reranking, structured-data queries, graph-based retrieval, source attribution, and permission-aware access controls.
  • Design knowledge-ingestion pipelines for documentation, runbooks, ticketing systems, configuration repositories, benchmark reports, experiment records, cluster state, telemetry catalogues, incident reports, and approved internal knowledge sources.
  • Build data and context pipelines combining unstructured knowledge with structured operational data including metrics, logs, traces, events, time-series databases, scheduler queues, job states, resource inventories, and configuration-management data.
  • Define tool contracts using structured input and output schemas, typed interfaces, validation, retries, idempotency controls, rate limits, timeouts, circuit breakers, approval requirements, and detailed audit logging.
  • Implement short-term task memory, long-term user or operational memory, episodic memory, and semantic memory with retention, access-control, and data-governance requirements.
  • Implement model-routing and fallback strategies across self-hosted inference endpoints and approved external models.
  • Partner with the Self-Hosted Inference Platform & Optimization team to ensure suitable endpoint profiles for agentic applications.
  • Build agent workflows for detection, diagnosis, recommendation, planning, action simulation, controlled execution, verification, and learning loops.
  • Develop offline replay, simulation, shadow-mode, and what-if evaluation capabilities to validate recommendations before allowing actions in production.
  • Design human approval and policy enforcement workflows presenting evidence, recommendation, expected impact, proposed action, confidence, risk classification, authorization scope, and rollback option.
  • Work with the Security Engineer to implement defense-in-depth controls against prompt injection, insecure output handling, excessive agency, unsafe tool use, data leakage, cross-tenant exposure, privilege escalation, credential misuse, unauthorized actions, and insufficient auditability.
  • Use policy engines, guardrail frameworks, structured output validation, content and tool filters, permission checks, sandboxing, and allowlisted action patterns.
  • Implement agent observability using tracing, metrics, logs, prompt and model version tracking, tool-call records, evaluation results, token and cost tracking, user feedback, incident evidence, and action audit trails.
  • Build and maintain evaluation frameworks using automated tests, curated test sets, simulation, replay, benchmark tasks, regression suites, model-based evaluators, human review, and operational-outcome measures.
  • Partner with the Kubernetes and custom scheduler team to consume and explain queue state, placement rationale, topology information, capacity signals, workload lifecycle events, policy outcomes, and performance data.
  • Partner with the Model-to-Grid product, inference, Platform, SDI, Security, UX, and global operations teams to turn agent capabilities into clear product workflows, production releases, runbooks, and measurable outcomes.

What we're looking for

  • 5+ years of software engineering experience, including 3+ years building AI/ML applications, distributed systems, automation platforms, data products, or production workflow systems.
  • Demonstrated experience delivering LLM-powered, agentic, retrieval-augmented, decision-support, or operational-automation applications into production.
  • Strong Python expertise, including experience with FastAPI or comparable API frameworks, asynchronous programming, event-driven services, distributed task execution, data pipelines, and API integrations.
  • Demonstrated ability to build custom agent orchestration when frameworks are insufficient, including state machines, graph-based execution, durable workflow execution, task queues, tool routers, model routers, planners, evaluators, and human-approval flows.
  • Experience with workflow and orchestration technologies such as Temporal, Dagster, Prefect, Airflow, Argo Workflows, Kubernetes Jobs, Celery, or equivalent event-driven or durable-execution platforms.
  • Strong understanding of multi-agent architectures, including supervisor-worker, planner-executor, router, reflection, critic-verifier, debate, hierarchical, blackboard, and event-driven coordination patterns.
  • Understanding of model protocols and integration approaches such as REST, gRPC, WebSockets, event streams, OpenAPI, JSON Schema, and Model Context Protocol (MCP) or comparable tool-integration patterns.
  • Experience integrating AI systems with time-series data, logs, traces, observability platforms, ticketing systems, configuration repositories, CI/CD systems, databases, cloud services, and enterprise APIs.
  • Familiarity with cloud-native AI platforms, including Kubernetes, containers, workload scheduling, model serving, GPU resources, observability, multi-tenancy, and operational runbooks.

Your team

AI & Applications

Reports to the Head of AI & Applications; partners with the Self-Hosted Inference Platform & Optimization team, Kubernetes and custom scheduler team, and Model-to-Grid product, inference, Platform, SDI, Security, UX, and global operations teams.

Apply

Save

About Firmus Technologies

Global leader pioneering the development and operation of efficient AI infrastructure across Asia Pacific

Firmus designs, builds, and operates a new class of digital infrastructure – the AI Factory. Through its model-to-grid technology approach, it has pushed the boundaries of multi-generational liquid cooling systems, energy management, AI software orchestration, and construction. Its large-scale GPU cloud platform, Firmus AI Cloud, delivers energy-efficient AI compute at scale to developers, enterprises, educational institutions, and government users.

Mission

To create the most efficient AI infrastructure by combining cutting-edge technology with a steadfast commitment to sustainability.

Key facts

  • 2019
  • Australia
  • AI infrastructure and cloud computing
  • Founded in Australia in 2019
  • NVIDIA Cloud and Engineering partner in Asia Pacific
  • Founder-led company
  • AI Factories designed to operate as assets to the energy grid

Partners & customers

NVIDIA

What they value

  • Sustainability
  • Ownership comes early
  • Minimum bureaucracy