AI and software for health and life sciences
We design and build AI in healthcare and life sciences for the work around patients and products: intake, documentation, quality records and the integrations that hold them together. Qualified people approve what matters, and data stays in the EU or in your own cloud.
Industries in health and life sciences
The work, and where software fits.
Hospitals, clinics, pharma companies and device manufacturers all run on documents and records that must be correct, traceable and kept for years. Much of the effort goes into reading, re-typing and cross-checking information between systems that were never designed to talk to each other.
Software fits where the rules are clear and the volume is high: sorting referrals, drafting records for review, searching controlled documents and moving data between the EPD, LIMS, eQMS and the tools around them. We build these with audit trails, evals and review steps, so clinicians, quality leads and regulatory teams keep the final say.
How we improve health and life sciences.
Document and referral intake
Referrals, lab reports, complaints and batch documents are read, structured and checked for completeness before a person confirms them.
Cited search over controlled knowledge
Protocols, SOPs and technical files become searchable in plain language, with each answer tied to the effective source document.
Clinical and quality system integrations
FHIR and HL7 links between EPDs and care apps, and data pipelines between LIMS, eQMS and ERP, so data is entered once.
Drafting with sign-off
Letters, deviation reports and complaint assessments drafted from existing records, reviewed and signed by the responsible professional.
Evidence for auditors and inspectors
Eval results, access logs and change records kept in a form that supports NEN 7510 audits, GxP validation and MDR technical files.
Related work.
Further reading.
Human-in-the-loop AI agents: design patterns for production
The patterns we use to run AI agents safely in production: approval checkpoints, confidence thresholds, tool permissions, audit trails and fallbacks.
4 min read
Document parsing with LLMs: extracting data from invoices and forms
How to turn invoices and forms into reliable structured data with LLMs, schemas, validation and a review queue for the cases that need a person.
4 min read
How to evaluate RAG: retrieval metrics, faithfulness and golden sets
How to measure a RAG system properly: separate retrieval from answers, check faithfulness claim by claim and build a golden set you can trust.
4 min read
Common questions.
Yes. We treat health data as a special category: we minimise and pseudonymise where we can, process in the EU or in your cloud, and document flows for your DPIA. Your data protection officer reviews the design before anything goes live.
No. Our builds extract, draft, search and flag, and a clinician, quality engineer or safety scientist makes the decision. Each approval is logged with the source material the software used.
Yes. We build against the interfaces those systems provide and coordinate access with the vendor and your IT department. We add software around core systems and leave the systems themselves in place.
With one process that has clear rules, a lot of manual reading or typing and a person who owns the outcome. Referral intake, SOP search and complaint triage are typical starting points, and our AI readiness checklist helps you pick.