AI Engineer (Agentic Systems)

AI Engineer (Agentic Systems)

Description

Build and productionize agentic AI workflows that help users prepare and analyze data, generate visualizations, extract insights, and synthesize outputs. Design how agents are orchestrated and how their interactions with tools, models, services, and data sources are made observable, testable, and reliable for production.

Focus on rebuilding an initial proof of concept into a robust internal analytical platform with strong production engineering practices.

Quick Facts

• Role focus: orchestration, observability, evaluation/testing, reliability, and production operations

• Product: internal analytical platform (web application + enterprise data layer + AI agents)

Role and Responsibilities

• Design agentic workflows for analysis, data preparation, visualization, insight extraction, and synthesis

• Own orchestration: how agents, tools, models, application services, and data sources interact

• Implement logging, tracing, diagnostics, and monitoring to understand what an agent did, why it did it, and where failures occurred

• Make AI behavior testable with evaluation approaches, regression tests, and structured test cases

• Design reliable interfaces using structured schemas and clear contracts between agents, tools, APIs, and application components

• Integrate with enterprise data by connecting to application APIs, databases, and analytical data platforms (not treating the LLM as the data-processing layer)

• Operate beyond prototypes: account for latency, cost, failure recovery, token usage, concurrency, model changes, security, and maintainability

• Implement guardrails for tool usage, data access, prompt injection, unexpected model behavior, and potentially destructive actions

• Collaborate with software and data engineers to keep the AI layer maintainable and product-aligned

• Contribute to architecture and product decisions through design reviews, sprint planning, and feedback cycles

• Use modern AI development tools such as Cursor, Claude Code, or equivalent agentic development environments

Requirements

• Strong Python development skills

• Hands-on experience building LLM-powered applications or agentic systems beyond simple prompt/API integrations

• Experience with an orchestration framework or equivalent architecture (e.g., LangChain, LangGraph, Google ADK, or comparable tooling)

• Strong understanding of agent orchestration, tool calling, state management, structured inputs/outputs, and failure handling

• Practical experience with logging, tracing, observability, and debugging for AI applications

• Experience evaluating and regression-testing agentic systems as models, prompts, tools, and workflows change

• Awareness of production AI concerns: security, permissions, prompt injection, token usage, latency, reliability, and cost

• Strong general software-engineering practices (version control, pull requests, automated tests, CI/CD, containerisation, clear architecture)

• Ability to work independently when some requirements are still being discovered

• Pragmatic approach to frameworks (choose what solves the problem well)

• Excellent communication skills and professional English fluency

• EU residency and permission to work in the EU

Benefits

• Build the AI layer of a real enterprise application from early stages

• Work alongside experienced engineers where reliability, traceability, and engineering quality matter as much as model capability

• Remote-first working culture and flexible working arrangements

• Challenging international projects and a highly experienced technical environment

• Fast-moving project with evolving priorities based on user interaction