Healthcare recruitment is one of the most time-sensitive industries there is. When a hospital needs to fill a shift, delays cost money and affect patient care.
Bluz is a recruitment platform for healthcare professionals, built by Campfire Digital using Flutter and Firebase. What started as a smart job board has evolved into something far more capable - an AI-powered recruitment assistant that removes friction at every stage of the process.
Here are three concrete ways AI has transformed the product.
Automated CV summarisation
Reviewing a CV has always been a slow, manual process. Healthcare CVs in particular tend to be dense - certifications, ward experience, shift preferences, specialisations.
We integrated Firebase's AI capabilities to automate this entirely. Whether a nurse uploads a scanned document, a PDF or a Word file, the system produces a clear, structured summary within seconds - experience, specialities, certifications and availability, all extracted and formatted consistently.
Recruiters get an actionable overview. Candidates get fair, consistent representation. Nobody wastes time on manual data entry.
Frictionless onboarding through data extraction
Signing up for a recruitment platform typically involves filling in form after form. For busy healthcare workers, this is a real barrier.
We solved it with an AI extraction layer. Upload your CV, and the system automatically pulls out:
- Name and contact details
- Location and travel range
- Field of practice and qualification level
- A profile summary
The profile is configured automatically. What used to take six or eight steps now takes two or three.
Natural language job postings with BluzBot
This is the feature that genuinely surprised people when they first saw it.
BluzBot lets hospital HR managers create job postings by typing naturally - no forms, no dropdowns. Something like:
"4x nurse - tomorrow night - surgery / 2x healthcare assistant - Tuesday day - geriatrics"
BluzBot handles the rest. It parses dates, roles, schedules and wards, retrieves the relevant pay rates, generates six separate postings with the correct metadata, and publishes them directly to the app - in seconds.
The technical architecture
The system works because Flutter and Firebase are tightly integrated:
- Callable Cloud Functions trigger AI operations the moment a user submits input
- Firestore triggers automate document processing when files land in storage
- Firebase Extensions enable rapid adoption of LLM models and document parsing
- Firebase Auth and Hosting provide security and scalability within Google Cloud
AI isn't bolted on as an afterthought here - it's woven into the architecture from the start.
What this means for product development
Traditional backend development is deterministic: structured data goes in, rules process it, structured output comes out. It works well for predictable tasks, but breaks down when faced with real-world ambiguity.
Generative AI flips this. Send a fragment of human intent - a rough sentence, a scanned document - and the model returns structured, ready-to-use data. The system adapts to how people actually communicate, rather than forcing people to communicate like machines.
For healthcare recruitment, where urgency and complexity are constant, this isn't just a technical improvement. It's a different way of thinking about what software can do.