Getting Started — Quickstart Guide

Get started with ByteEngine in under 15 minutes — from account creation to running your first healthcare AI workflow.

Overview

ByteEngine is built to make healthcare AI orchestration and FHIR data automation easy — even if you've never worked with AI or healthcare standards before.

In this Quickstart, you'll learn how to:

• Create a FHIR or DICOM data store
• Add and view healthcare data
• Deploy an AI Worker (agent)
• Create an automated workflow
• View AI results and logs

Goal: Build a workflow that summarizes new patient observations automatically using a healthcare AI agent.

Step 1: Sign Up & Log In

1. Go to ByteEngine Console
2. Click Sign Up
3. Enter your email and password or sign in with Google

UI Placeholder: [Screenshot: ByteEngine Signup Page showing region selector and compliance info.]

Step 2: Create Your First FHIR Server

1. Navigate to HealthDataStore → Create New Server
2. Select FHIR Server
3. Enter:
   • Name: demo-fhir-server
   • Version: R4
   • Region: US
4. Click Deploy

In a few seconds, your FHIR server is live!

UI Placeholder: [Screenshot: FHIR server dashboard with base URL and endpoint list.]

Example API Endpoint

Your server URL will look like this:

https://api.engine.boolbyte.com/api/fhirserver/{server_id}

Using JavaScript SDK

import { EngineClient } from '@boolbyte/engine';

const client = new EngineClient({ apiKey: 'YOUR_API_KEY' });

// Create a FHIR server
const fhirServer = await client.dataStore.createFhirStore({
  name: 'demo-fhir-server',
  region: 'global',
  fhirVersion: 'R4',
  type: 'serverless'
});

console.log('FHIR Server created:', fhirServer.data.endpoint);

Step 3: Add Sample FHIR Data

Now let's add a sample Patient and an Observation.

Example: Create a Patient

curl -X POST https://api.engine.boolbyte.com/api/fhirserver/{server_id}/Patient \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/fhir+json" \
  -d '{
    "resourceType": "Patient",
    "name": [{ "given": ["John"], "family": "Doe" }],
    "gender": "male",
    "birthDate": "1980-03-10"
  }'

Example: Create an Observation

curl -X POST https://api.engine.boolbyte.com/api/fhirserver/{server_id}/Observation \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/fhir+json" \
  -d '{
    "resourceType": "Observation",
    "status": "final",
    "code": { "text": "Oxygen Saturation" },
    "subject": { "reference": "Patient/1" },
    "valueQuantity": { "value": 95, "unit": "%" }
  }'

UI Placeholder: [Screenshot: FHIR Explorer view showing Patient and Observation entries.]

Using JavaScript SDK

// Initialize FHIR client
await client.dataStore.initializeFhirStoreClient(fhirServer.data.id);
const fhirClient = client.dataStore.getFhirStoreClient();

// Create a patient
const patient = await fhirClient.create({
  resource: {
    resourceType: "Patient",
    name: [{ given: ["John"], family: "Doe" }],
    gender: "male",
    birthDate: "1980-03-10"
  }
});

// Create an observation
const observation = await fhirClient.create({
  resource: {
    resourceType: "Observation",
    status: "final",
    code: { text: "Oxygen Saturation" },
    subject: { reference: `Patient/${patient.id}` },
    valueQuantity: { value: 95, unit: "%" }
  }
});

console.log('Patient created:', patient.id);
console.log('Observation created:', observation.id);

Step 4: Create an AI Worker

AI Workers are intelligent agents that reason, act, and interact with your FHIR data.

1. Go to AI Workers → New Worker
2. Choose a Model:
   • medgemma-27b (healthcare-optimized)
   • gpt-oss-120b (general-purpose)
3. Set Worker Name: VitalsSummarizer
4. Add Instructions:

   Summarize all new Observation records for each Patient and highlight significant changes.

5. Save and Deploy

UI Placeholder: [Screenshot: Worker creation form with instruction editor and model selection.]

Using JavaScript SDK

// Create an AI Worker
const worker = await client.worker.createWorker({
  name: 'VitalsSummarizer',
  description: 'AI agent that summarizes patient vital signs and observations',
  defaultModelName: 'medgemma-27b',
  instructions: 'Summarize all new Observation records for each Patient and highlight significant changes.',
  toolConfigs: {
    tools: [
      {
        toolName: 'fhir_access',
        config: { serverId: fhirServer.data.id }
      }
    ]
  }
});

console.log('AI Worker created:', worker.data.id);

Step 5: Create a Workflow

Now, let's connect everything together.

1. Go to Workflows → New Workflow
2. Drag:
   • FHIR Server (your demo server)
   • AI Worker (VitalsSummarizer)
3. Add a Trigger:
   • Event: on Observation.created
4. Click Run Workflow

UI Placeholder: [Screenshot: Visual Workflow Builder showing connections between FHIR Server and AI Worker.]

YAML Configuration Example

workflow:
  name: summarize-new-observations
  trigger:
    event: Observation.created
    source: demo-fhir-server
  steps:
    - type: worker
      name: summarize-vitals
      worker: VitalsSummarizer

Using JavaScript SDK

// Create a workflow
const workflow = await client.workflow.createWorkflow({
  name: 'summarize-new-observations',
  trigger: {
    type: 'fhir',
    event: 'Observation.created',
    source: fhirServer.data.id
  },
  steps: [
    {
      id: 'summarize-vitals',
      type: 'worker',
      worker: worker.data.id,
      input: '{{trigger.resource}}'
    }
  ]
});

console.log('Workflow created:', workflow.data.id);

Step 6: View AI Results

After your workflow runs, go to the Sessions tab. You'll see the Worker's reasoning and output:

{
  "session_id": "abc123",
  "worker": "VitalsSummarizer",
  "response": "Patient John Doe shows stable oxygen levels at 95%. No concerning changes detected."
}

UI Placeholder: [Screenshot: Session log view with AI output and FHIR resource reference.]

Using JavaScript SDK

// Create a session and run a task
const session = await client.session.createSession({
  workerId: worker.data.id,
  metadata: { context: 'vital signs analysis' }
});

const task = await client.task.createTask(session.data.id, {
  instructions: 'Analyze the latest observation for Patient John Doe and provide a summary',
  model: 'medgemma-27b'
});

console.log('Task completed:', task.data);

Step 7: Explore More

Now that you've built your first AI workflow, explore what else ByteEngine can do:

Feature	Description	Example
Pipelines	Connect legacy databases and map to FHIR using AI	Auto-convert SQL schema to FHIR
Subscriptions	Trigger bots on new FHIR events	Notify when new lab results are created
Knowledge Bases	Store patient notes for RAG-powered chatbots	"Show me all recent radiology notes"
File Storage	Store DICOM or PDF files alongside FHIR	Upload and link patient scans

Next Steps

• Full API Documentation →
• Learn About AI Workers →
• Build Custom Workflows →
• Developer SDKs →
• Knowledge Bases →

Summary

Step	Action	Outcome
1	Sign up and choose data residency	HIPAA/GDPR-compliant workspace
2	Create FHIR server	Live FHIR API endpoint
3	Add sample data	Patient and Observation records
4	Create AI Worker	Automated summarization model
5	Create Workflow	Connect data and AI logic
6	View Results	AI-generated clinical insights

You just built a healthcare AI workflow!

ByteEngine automates healthcare data orchestration — so you can focus on innovation, not infrastructure.

Build Something Amazing

Ready to build your healthcare application? Here are some ideas:

• Clinical Decision Support - AI Workers that analyze patient data and suggest treatments
• Administrative Automation - Agents that handle scheduling, billing, and documentation
• Research Analytics - Tools that extract insights from large FHIR datasets
• Patient Engagement - SMART Apps that help patients manage their care

Join our community to connect with other healthcare innovators!