AI Engineer: Agentic Systems Orchestration

Aether Biomedical; United States; Remote; USD 140,000 - 190,000; Full time

Benefits offered by this job

Remote-first culture

Flexible working arrangements

Challenging international projects

Experienced technical environment

Competitive compensation

Job summary

PortBlueSky is seeking an AI Engineer to advance an internal analytical platform that combines a modern web UI with enterprise data for AI agents to analyse data, create visualisations, extract insights and synthesize outputs. You will shape agent orchestration, tracing, evaluation and operation in production, with language-agnostic tooling and strong production practices.

You’ll work with Python, FastAPI and LangChain/LangGraph, connecting to databases and APIs while ensuring performance,

Qualifications

• Strong Python development skills.
• Hands-on experience building LLM-powered applications or agentic systems.
• Experience with orchestration frameworks (LangChain, LangGraph, Google ADK) or similar.
• Understanding of agent orchestration, tool calling, state management, inputs/outputs, and failure handling.
• Experience with logging, tracing, observability, and debugging for AI apps.
• Ability to evaluate/regression-test agentic systems as models and tools evolve.
• Awareness of production AI concerns: security, prompts, latency, cost.
• Software-engineering practices: VCS, PRs, automated tests, CI/CD, containers.
• Ability to work independently in evolving requirements.
• Clear, pragmatic approach to technology choices.

Responsibilities

• Design agentic workflows: Build agents and multi-step AI workflows for analysis, data prep, visualization, insight extraction, and synthesis.
• Own orchestration: Design interactions between agents, tools, models, services, and data sources.
• Build for observability: Implement logging, tracing, diagnostics, and monitoring to understand agent actions and failures.
• Make AI behaviour testable: Develop evaluation approaches and regression test cases for evolving prompts and tools.
• Design reliable interfaces: Use schemas and clear contracts between agents, tools, APIs, and components.
• Integrate with enterprise data: Connect AI workflows to APIs, databases, and data platforms without treating LLM as data layer.
• Operate beyond prototypes: Consider latency, cost, failure recovery, token usage, concurrency, and security from day one.
• Build appropriate guardrails: Implement controls around tool usage, data access, prompt injection, and risks.
• Collaborate with engineers: Ensure AI layer is a maintainable part of the product, not an isolated experiment.
• Contribute to architecture and product decisions: Engage in design reviews, planning, demos, and feedback.
• Use modern AI tools: Work with Cursor, Claude Code, or equivalent agentic environments.

Skills

Python

LLM-powered apps

Agent orchestration

Observability

Testing & regression

Security & reliability

CI/CD

English comms

Independent work

Framework flexibility

Tools

Docker

PostgreSQL

AWS

FastAPI

GitHub Actions

Kubernetes

LangChain

LangGraph

Google ADK

Pydantic

OpenTelemetry

Job description

PortBlueSky is seeking an AI Engineer to advance an internal analytical platform that combines a modern web UI with enterprise data for AI agents to analyse data, create visualisations, extract insights and synthesize outputs. You will shape agent orchestration, tracing, evaluation and operation in production, with language-agnostic tooling and strong production practices.

You’ll work with Python, FastAPI and LangChain/LangGraph, connecting to databases and APIs while ensuring performance,