Software Engineer

We are looking for a Software Engineer who can design the data and AI systems behind that capability. This is not an AI-for-AI's-sake role. You will choose between deterministic software and large language models according to the problem, ensuring every important output is explainable, testable and grounded in approved engineering knowledge.

You will help shape a new internal software capability, work on real engineering problems and influence architecture, standards and delivery practices from an early stage. Your work will be used by engineers across Europe, the Middle East, Asia and Australia.

ROLE RESPONSIBILITIES

• Build Python and Fast

API services that convert engineering inputs into validated, traceable outputs.

• Design versioned schemas, PostgreSQL data models, migrations and validation using Pydantic or JSON Schema.
• Extract structured knowledge from standards, spreadsheets, templates and drawings, using pandas, openpyxl and LLM APIs where appropriate.
• Create evaluations, representative test datasets and automated checks that make AI-supported workflows measurable and safe.
• Implement access controls, audit logs and safeguards for confidential company and client information.
• Deploy maintainable services on Azure, GitHub, and others with practical monitoring and failure handling.
• Work directly with discipline engineers to understand intent, resolve knowledge gaps and demonstrate progress every two weeks.
• Review code and help establish the technical standards for a growing internal software capability.

JOB REQUIREMENTS

• Bachelor’s degree in Computer Science, Software Engineering, Information Technology, Engineering, or a related field.
• Equivalent practical experience in software engineering may also be considered.
• At least 3 years of professional software development, including Python.
• Strong relational data modelling and SQL, including schemas, migrations and evolving data structures.
• Experience building APIs, handling authentication and designing for imperfect third-party behaviour.
• Practical experience integrating LLM APIs, including structured outputs and evaluation - not only using chat assistants.
• Automated testing with pytest or similar, plus Git and peer code review as normal practice.
• Sound judgement when using AI coding assistants: generated code is a draft to verify, not an answer to accept.
• Clear written English for specifications, documentation and asynchronous collaboration across regions.

.

Helpful, but not essential

• Calculation or rules engines, regulated systems, or other auditable and reproducible domains.
• React or Next.js, Azure, Docker, GitHub-based CI and Model Context Protocol integrations.