Vida

Vida presents its product as a catalog of AI employees: receptionist, appointment scheduler, lead qualification specialist, call recovery specialist, customer support representative, intake coordinator and administrative assistant. Each is a configured agent with instructions, business knowledge, a voice and a set of connected tools, and the vendor describes the role framing plainly: "An AI employee is an AI agent configured for a business role, such as receptionist, appointment scheduler, or customer support representative."

The conversational side covers phone, text and email, with concurrent conversations included on every plan. The second half of the platform is the Computer Agent, which works in software instead of talking to customers. Vida's documentation splits the two deliberately: a conversational agent keeps the caller, then hands a bounded job to a Computer Agent that uses skills, a browser session, workspace files and memory, and returns a linked work conversation showing its tool activity and result.

The plan table starts at Starter, $250 a month for one agent with no included usage. Business 500, 1k and 2k cost $700, $1,150 and $2,050 a month and prepay $500, $1,000 and $2,000 of monthly usage, which Vida frames as saving $600, $1,200 and $2,400 a year. The allowance is a dollar figure shared across every agent in the organization and does not roll over.

Usage is the part that moves. Voice is $0.15 a minute billed in 60-second increments, texts are $0.06 a message with no fee on inbound, and AI tokens have their own rate card. Additional agents are $250 each a month, phone numbers $5, and text messaging a $25 monthly add-on that also needs carrier registration Vida says can take up to two weeks. Outbound calling and managed agents are per-agent add-ons with no published price, and the terms add a Fair Use Policy over tokens, minutes and messages. There is no plan or platform setup fee, but one-time setup fees for selected agents and add-ons appear in the quote.

The Vida API covers agent configuration with staging and publish, saved versions and weighted experiments, outbound call, SMS and email task queues of up to 10,000 targets, contacts and contact lists, conversation logs with CSV export and computed metrics, webhook destinations and relays, SIP registration and routing, and the full Computer Agent lifecycle including skills, secrets, workspace files and browser automation sessions.

Vida also publishes a machine-readable operating guide, the API Skill, sourced from a public repository, for outside AI agents configuring Vida through the API. That is a documentation artifact, not an access path: "The skill does not grant access to Vida. Every request still requires a Vida API token with permission for the selected account." No MCP server is documented, and the integration comparison table lists apps, built-in functions, webhook functions, event webhooks, Zapier, reusable helpers, Computer Agents and Canvas as the available approaches.

The operator is VIDA Global Inc., registered in Texas with an address at 12160 W. Parmer Lane, Suite 130-716, Cedar Park, TX 78613, named in both the privacy policy and the terms. Press logos on the home page cite Forbes, Fast Company, Entrepreneur and TechRadar.

A large share of the platform exists for partners rather than end businesses. The docs carry a reseller and partner account hierarchy, agent templates with migrations, white-label domains, an embeddable customer application, Stripe and Chargebee billing for downstream customers, agent detail record exports to an external rating engine, and reseller payouts. A buyer evaluating Vida for one business will meet features gated to "eligible reseller" accounts throughout the documentation.

Vida states it undergoes regular independent SOC 2 Type II audits covering security, availability and confidentiality, with the report on request, and supports HIPAA-regulated workflows through designated configurations and a BAA. The data addendum is specific where it counts: customer content is not used to train general-purpose models without written authorization, and model, speech-to-text and text-to-speech providers are contractually held to the same restriction. Security incidents are notified within 72 hours, and encryption in transit is committed with encryption at rest "where supported by the applicable systems".

Two limits sit beside that. All data is stored in the United States, with Standard Contractual Clauses for EU and UK transfers, so there is no region to pick. And the pricing page sells a "SOC 2 Security" add-on and an "Advanced Privacy" add-on, both quoted by sales, with the FAQ saying HIPAA workflows "require the appropriate Advanced Privacy setup and authorization" before PHI is handled. The company publishes no SLA: the terms say "We cannot guarantee the Services will be available at all times."

The home page FAQ answers "Is Vida HIPAA compliant?" with "Yes, Vida supports HIPAA-regulated use cases through designated HIPAA-enabled configurations." The pricing page FAQ drops the yes and adds a condition: "Vida supports HIPAA-regulated workflows through designated configurations and an applicable Business Associate Agreement (BAA). These workflows require the appropriate Advanced Privacy setup and authorization before handling protected health information." The second is the one to plan around, since Advanced Privacy is a quoted add-on.

Self-service is similarly two-sided. The start page asks only for a work email, while the FAQ describes a guided path: "Our team will help you choose an AI employee and review the workflow, systems, and setup you need." No page documents single sign-on through an identity provider, and no on-premise deployment is offered.

The fit is an appointment-driven business losing work when nobody picks up: the industry pages name home services, health clinics, telehealth, legal, property management and auto dealerships, and the published outcome metrics are confirmation rate, appointments and minutes saved. Existing numbers can be forwarded to a Vida agent for after-hours or overflow cover, though the FAQ notes forwarding calls does not forward texts.

It fits less well in three cases. A buyer wanting to test before committing has no free plan and no trial, only a demo call. A buyer who needs predictable monthly cost faces token, minute and message metering on top of the subscription, with AI token rates on a separate card. And a team that needs SSO, a region outside the United States, or a contractual uptime figure will not find any of them published.