You'll build and operate the shared AI capabilities behind Aidn's digital colleague direction.
Creating the reusable components, APIs, evaluation patterns, and tooling that let other product teams adopt AI safely, without reinventing core infrastructure.
Taking capabilities like speech-to-summary, form-filling, and contextual drafting from pilot to production, tied to the correct patient, case, and data context, with real attention to latency, cost, reliability, and failure modes.
Building safe execution into the platform: previews before actions, human-in-the-loop checkpoints, risk tagging, rollback, full traceability and audit trails, and the separation between MDR and non-MDR contexts.
Establishing how we test, measure, and monitor AI behaviour over time, eval frameworks, ground-truth validation, regression detection, and production monitoring, so quality and trust hold as usage scales.

Strong software and systems engineering: you've built and operated production systems, with sound judgment on reliability, scalability, and maintainability.
Shipped applied AI in production, not just prototypes: you've taken LLM
or speech-based capabilities into real use and handled latency, cost, reliability, and failure modes.
Built things other engineers depend on: shared services, APIs, libraries, or platform components, with the interfaces and documentation that made adoption easy.
Evaluation and observability as a habit: you measure system behaviour and build confidence through evidence, not assumption.
Safe-by-design engineering: comfortable building with guardrails, traceability, audit trails, human-in-the-loop checkpoints, and controlled execution.
Strong communication and clear technical writing in English.

Aidn

Open Positions (1)