Glossary

Technical terms, explained plainly

Healthcare IT, AI architecture, e-commerce, and SaaS concepts — defined for founders who need to understand them, not just say them.

20 terms definedHealthcare, AI, E-commerce, SaaS
ABCEFHILMPRSV
Healthcare

ADT Feed

Admit, Discharge, Transfer. An ADT feed is a real-time data stream that notifies connected systems whenever a patient is admitted, discharged, or transferred within a facility. ADT events are transmitted in HL7 format and are the backbone of care coordination and EMR integration. A home health platform receiving ADT events from a hospital knows in real time when one of its patients is admitted — enabling proactive outreach, care plan updates, and staffing adjustments. Without ADT integration, home health agencies often learn about patient events days later, typically from a phone call.

Healthcare Software DevelopmentCareCoordinations Case Study
AI & Automation

AI Agent

An AI agent is a software system that uses a large language model to reason, plan, and take actions autonomously — not just respond to a single prompt. Unlike a standard chatbot that answers one question at a time, an AI agent can break a complex goal into sub-tasks, call external tools (APIs, databases, search), evaluate results, and iterate until the goal is achieved. Agents built with frameworks like LangChain or CrewAI can draft emails, query databases, summarize documents, trigger workflows, and self-correct when a step fails. The key property is goal-directed autonomy: given a high-level objective, the agent figures out the steps.

AI & Automation Services
Healthcare

BAA

Business Associate Agreement. Under HIPAA, a BAA is a legally required contract between a covered entity (hospital, clinic, health plan) and any business associate that handles Protected Health Information (PHI) on its behalf. If a software vendor stores, processes, or transmits PHI — which includes almost any healthcare SaaS product — HIPAA requires a signed BAA before that vendor can access the data. The BAA specifies how PHI must be safeguarded, what happens in the event of a breach, and how PHI must be destroyed at the end of the engagement. A vendor that cannot or will not sign a BAA cannot legally handle PHI.

Healthcare Software Development
Healthcare

Care Coordination

Care coordination is the deliberate organization of patient care activities across multiple providers, care settings, and time periods. In home health, this means ensuring that a patient's primary physician, home health agency, DME supplier, and hospital are all working from the same plan — updated in real time as the patient's condition changes. Software supporting care coordination typically handles ADT event notification, care plan management, IDG meetings, task assignment and completion tracking, and communication logs. Poor care coordination is one of the primary causes of preventable hospital readmissions in the United States.

CareCoordinations Case StudyHealthcare Software Development
Healthcare

EHR

Electronic Health Record. An EHR is a digital record of a patient's health information that can be shared across different healthcare settings. Unlike an EMR (Electronic Medical Record), which is typically confined to a single practice, an EHR is designed for interoperability — it can be accessed and updated by multiple providers involved in a patient's care. Major EHR systems include Epic, Cerner (now Oracle Health), athenahealth, and HCHB (for home health). Integrating with an EHR requires understanding the vendor's API, authentication model, and data formats — which are typically HL7 v2 or FHIR.

Healthcare Software Development
Healthcare

EMR

Electronic Medical Record. An EMR is the digital version of a patient's chart within a single healthcare practice or organization. It contains the medical and treatment history of patients within one practice — diagnoses, medications, allergies, lab results, and visit notes. The key distinction from an EHR is scope: an EMR stays within the practice, while an EHR is designed to move with the patient across providers. For software integration purposes, EMR and EHR are often used interchangeably, and the technical integration approach — HL7, FHIR APIs, or proprietary SDKs — is the same regardless of which term the vendor uses.

Healthcare Software Development
Healthcare

FHIR

Fast Healthcare Interoperability Resources. FHIR (pronounced "fire") is a modern standard for exchanging healthcare data electronically, developed by HL7 International. Unlike older HL7 v2 messages, FHIR uses REST APIs, JSON or XML, and a standardized set of data "resources" (Patient, Encounter, Observation, Medication) that any compliant system can understand. The US 21st Century Cures Act mandates FHIR API access for most EHR systems, making it the dominant standard for new healthcare integrations. Building FHIR-based integrations is significantly faster than legacy HL7 v2 because the interfaces are web-standard and well-documented.

Healthcare Software Development
Healthcare

HCHB

Homecare Homebase. HCHB is the dominant EMR and EHR platform for home health and hospice agencies in the United States. It handles clinical documentation, scheduling, billing, and compliance for field clinicians who visit patients at home. Integrating with HCHB requires working with their proprietary API and data model, which differs from standard FHIR. We have built bidirectional HCHB integration in production, including ADT event feeds, MDM attachment synchronization, and real-time two-way data sync — making it one of the more complex but highest-value healthcare integrations we offer.

CareCoordinations Case StudyHealthcare Software Development
Healthcare

HIPAA

Health Insurance Portability and Accountability Act. HIPAA is a US federal law enacted in 1996 that establishes national standards for protecting sensitive patient health information. For software developers, HIPAA compliance means implementing technical safeguards: encryption of Protected Health Information (PHI) in transit and at rest, role-based access control, audit logging on every PHI access, automatic session timeouts, and BAA agreements with all vendors who handle PHI. Non-compliance carries fines from $100 to $50,000 per violation. HIPAA compliance is an architectural decision, not a checkbox — it must be built in from day one, not retrofitted.

Healthcare Software Development
Healthcare

HL7

Health Level Seven. HL7 is a set of international standards for the exchange, integration, sharing, and retrieval of electronic health information. HL7 v2 — the older, pipe-delimited message format — is still the most widely used standard in hospital systems for sending ADT events, lab results, and clinical documents between systems. HL7 v3 was largely bypassed in favor of FHIR, which is HL7's modern REST API standard. Implementing HL7 v2 integration requires parsing the pipe-delimited message format, understanding segment structures (MSH, PID, PV1, OBR), and handling considerable variation in how different vendors implement the "standard."

Healthcare Software Development
Healthcare

IDG Meeting

Interdisciplinary Group Meeting. In home health, an IDG meeting (also called an interdisciplinary team or IDT meeting) is a required clinical review in which the full care team — nurse, physical therapist, occupational therapist, social worker, and physician — reviews each patient's care plan, discusses progress, and updates goals. CMS requires IDG meetings at least every 14 days for hospice patients and at clinical milestones for home health patients. Software supporting IDG meetings must handle attendance tracking, care plan versioning, meeting minutes, task assignments, and regulatory compliance documentation — all tied to individual patient records.

CareCoordinations Case StudyHealthcare Software Development
AI & Automation

LangChain

LangChain is an open-source framework for building applications powered by large language models. It provides abstractions for chaining LLM calls, connecting models to external tools (APIs, databases, search engines), implementing RAG pipelines, and building AI agents that can reason and act across multiple steps. LangChain supports multiple LLM providers (OpenAI, Anthropic, Google) and vector databases (Pinecone, Chroma, Weaviate), making it the most widely used framework for production AI integrations. Alternatives include LlamaIndex (better for document-heavy RAG) and CrewAI (better for multi-agent orchestration).

AI & Automation ServicesRAG Pipeline Case Study
AI & Automation

LLM

Large Language Model. An LLM is a type of AI model trained on massive text datasets to understand and generate human language. Modern LLMs like GPT-4o, Claude, and Gemini can answer questions, write code, summarize documents, classify content, and reason through complex problems. For product development, LLMs are most valuable when they can access your specific data — which requires integration patterns like RAG (Retrieval-Augmented Generation). A raw LLM without grounding will generate plausible-sounding but potentially incorrect answers; a grounded LLM with RAG answers from your actual content with traceable sources.

AI & Automation Services
Web & SaaS

Multi-tenant SaaS

A multi-tenant SaaS application serves multiple customers (tenants) from a single shared infrastructure, while keeping each tenant's data isolated from the others. Tenants may access the platform via separate subdomains (customer1.app.com) or via a single domain with tenant identification through URL path or auth token. Key technical decisions in multi-tenant architecture include tenant isolation strategy (shared database with tenant ID, separate schemas, or separate databases), custom domain support, per-tenant feature flags, role-based access control scoped to each tenant, and usage metering for per-seat or usage-based billing via Stripe.

Web & SaaS Development
Healthcare

PHI

Protected Health Information. Under HIPAA, PHI is any information that can identify an individual and relates to their health condition, healthcare treatment, or payment for healthcare. This includes names, dates (except year), geographic identifiers, phone numbers, email addresses, medical record numbers, health plan numbers, and biometric data — when combined with health information. PHI includes both structured data (database records) and unstructured data (doctor's notes, images, audio recordings). Any software that stores, processes, or transmits PHI must implement HIPAA safeguards, and the vendor must sign a Business Associate Agreement (BAA).

Healthcare Software Development
AI & Automation

RAG Pipeline

Retrieval-Augmented Generation. A RAG pipeline is the architecture that makes AI answer from your actual data instead of hallucinating. It works in three stages: first, documents are chunked and converted into vector embeddings stored in a vector database (Pinecone, Chroma, Weaviate). Second, when a user asks a question, the question is also converted into a vector and the most semantically similar document chunks are retrieved. Third, those retrieved chunks are passed to an LLM as context, and the model generates an answer grounded in your real content. RAG is the standard architecture for knowledge assistants, internal Q&A tools, and customer support bots that must answer from a specific document set.

AI & Automation ServicesRAG Pipeline Case Study
Mobile

React Native

React Native is a cross-platform mobile development framework created by Meta that lets developers write a single JavaScript codebase that compiles to native iOS and Android apps. Unlike web-based hybrid approaches (Cordova, Ionic), React Native renders actual native UI components — not a web view — which gives it near-native performance and access to the full native API surface. For most product teams, React Native reduces mobile development cost by 40–60% versus maintaining separate native codebases, with acceptable trade-offs in performance-sensitive scenarios. The main alternative is Flutter (Dart-based, by Google), which offers slightly better rendering consistency but a smaller ecosystem.

Mobile App Development
E-commerce

Shopify Plus

Shopify Plus is the enterprise tier of the Shopify e-commerce platform, designed for high-volume merchants doing $1M+ in annual revenue. It unlocks features unavailable on standard Shopify: custom checkout with Checkout Extensibility, multiple storefronts under one account, wholesale and B2B channels, higher API rate limits, and advanced automation via Shopify Flow. For development teams, Shopify Plus opens access to the Checkout API and allows more aggressive customization of the purchase funnel. The decision to build on Shopify Plus versus custom typically hinges on whether your business logic can be expressed within Shopify's data model — if it cannot, custom is the right answer.

E-commerce Development
Engagement

Staff Augmentation

Staff augmentation is an engagement model where an outside vendor provides individual engineers who work directly within the client's team — attending standups, using the client's tools, and operating under the client's processes — rather than working as a separate project team. The client retains full control over sprint planning, architecture decisions, and day-to-day management. Staff augmentation is best suited for teams that need to fill specific skill gaps quickly, scale up for a defined period, or move faster on a backlog without the overhead of hiring. At Nexios, staff augmentation is $20/hour with a minimum of 80 hours per month.

Services Overview
AI & Automation

Vector Database

A vector database is a specialized data store designed to store and query high-dimensional vector embeddings efficiently. In AI applications, text, images, and other content are converted into numerical vectors (embeddings) by a model like OpenAI's text-embedding-3-large. These vectors encode semantic meaning, so similar content has similar vectors. A vector database retrieves the most semantically similar items to a query in milliseconds, even across millions of documents. This retrieval step is the R in RAG — Retrieval-Augmented Generation. Common vector databases include Pinecone (managed SaaS), Weaviate (open-source, self-hostable), Chroma (local development), and pgvector (PostgreSQL extension).

AI & Automation ServicesRAG Pipeline Case Study

Let's talk

Have a project in mind?

Tell us about your project below, or pick another way to reach us. Average response time: under 4 business hours.