AI Solution Architect (Document AI, Contract)

Intelance is a specialist architecture and AI consultancy working with clients in regulated, high-trust environments (healthcare, pharma, life sciences, financial services). We are building a small senior team to deliver an AI-assisted clinical tool for a UK-based organisation in human genetic testing. This is a contract / freelance role where you will act as the lead AI Solution Architect on a real production system . Tasks • Own the end-to-end solution architecture for an AI-assisted tool that reads clinical-style reports (PDF/Word), applies scoring rules, and returns explainable outputs into an existing assessment platform. • Translate business goals, marking rubrics, and regulatory constraints into a concrete technical blueprint (data flows, components, interfaces, security boundaries). • Design a hybrid scoring approach that combines deterministic rules with LLM-powered NLP, including confidence scoring, thresholds, and human-in-the-loop workflows. • Define the data ingestion pipeline together with a Data Engineer (OCR, parsing, normalisation, metadata, lineage, logging). • Shape the integration pattern with the client’s existing QA/assessment system (API, CSV import/export, SFTP drop-zone), ensuring robustness and traceability. • Work closely with the ML Engineer to choose modelling approaches, evaluation metrics, and explainability patterns that are realistic for a regulated environment. • Collaborate with a Compliance/Validation lead to ensure the design supports GDPR, UK data protection, and GxP/UKAS-style validation requirements (audit trails, evidence, access control). • Produce clear design artefacts: architecture diagrams, interface specs, data schemas, non-functional requirements, and decision logs. • Support the Delivery Lead/PM with estimates, risk identification, and delivery planning across discovery, pilot, and integration phases. • Join client workshops (remote, UK hours) to present options, explain trade-offs, and help non-technical stakeholders understand how the solution will work in practice. Requirements Must-have • 6+ years in Solution Architecture / Data Architecture / AI Architecture with at least one end-to-end delivery of a data-intensive or ML-enabled system. • Strong hands-on background with cloud (ideally Azure), APIs, and modern data / AI stacks (e.g. Python, containers, message queues, object storage). • Practical experience designing solutions involving NLP or document processing (e.g. PDF/Word ingestion, OCR, text extraction, classification, or scoring). • Good working knowledge of LLMs / generative AI and how to combine them with rules, prompts, and evaluation frameworks in production. • Ability to design for security, privacy, and auditability: data minimisation, encryption, logging, RBAC, basic DPIA thinking. • Comfortable working in or around regulated domains (healthcare, life sciences, pharma, financial services, or similar). • Strong communication skills – able to talk to both engineers and senior non-technical stakeholders, and to document decisions clearly. • Available for 2-3 days per week on a contract/freelance basis, working largely remotely in UK or close European time zones. Nice-to-have • Experience with Azure AI / OpenAI, Azure Functions, Event-driven architectures, or any major cloud document-AI/OCR service. • Previous work with UK NHS, UKAS, ISO 15189, GxP, or other lab/clinical quality frameworks. • Experience integrating with legacy or “closed” systems via CSV, SFTP or intermediate staging services. • Prior consulting or client-facing experience in small, senior teams. Benefits • High-impact project: build a production AI tool used by laboratories globally to support external quality assessment in human genetics. • Senior, lean team: work directly with the leadership team and other experienced consultants – no bureaucracy, fast decisions. • Remote-first, flexible: work from anywhere compatible with UK business hours; engagement planned at 2–3 days/week initially. • Contract / freelance: competitive day rate, potential extension into a multi-year roadmap across additional schemes and products. • Opportunity to shape reusable AI building blocks (architecture patterns, components) that will be used across future Intelance engagements. We review every application personally. If there’s a fit, we’ll invite you to a short intro call to walk through the project context and next steps.