Pasito
AI-native workspace for group benefits: extraction agents turn plan documents, contracts and proposals into a normalized source of truth, then generate the guides, microsites, comparisons and campaigns for a plan year, with a 24/7 benefits assistant answering members from the employer's own plan documents.
Pasito is an AI-native workspace for the group benefits industry, used by insurance carriers, benefits consultants and employers to replace the manual document work that surrounds every plan year. It was founded in 2021 in New York by Pauline Roteta and raised a Series A led by Insight Partners in February 2026.
The problem it starts from is unstructured paper. Plan documents, contracts, broker proposals and census files arrive in whatever format they were written in, and someone has to read them and rebuild the plan by hand. Pasito's agents scan those documents, infer the schema, and standardize the contents into a single source of truth that everything else is generated from. Proposal comparison turns a stack of RFP responses into a side-by-side view in seconds; RFP intake management structures them into something navigable.
From that layer the platform produces the assets a plan year needs: benefits guides, highlight sheets, enrollment presentations, marketing documents and personalized plan recommendations, each templatized and generated rather than rewritten. Campaigns go out over email and SMS and link members straight into a benefits microsite and decision support, which leads with a personalized recommendation rather than a long intake form. An AI benefits assistant answers coverage questions 24/7 in the client's own brand, with administrators controlling its tone, its instructions and which plan documents it may answer from. Agents can be deployed fully white-labeled.
Because the domain is regulated and fiduciary, the compliance posture is unusually detailed for a company this size: SOC 2 Type II independently audited by Prescient Assurance with the report available on request, HIPAA safeguards with Business Associate Agreements, role-based access control with least privilege and revocation protocols, per-client data isolation, and a published AI governance framework covering pre-deployment testing and post-launch evaluation. Pasito holds a Zero Data Retention Agreement with Anthropic, so inputs and outputs are processed ephemerally and discarded when a session ends. Pricing is quote-only.
Vendor details
Canonical URL
https://pasito.ai
Category
Enterprise operations agent
Subcategory
Employee benefits and insurance operations
Funding status
Independent, headquartered in New York, founded in 2021 by Pauline Roteta, a former BlackRock executive, and a Y Combinator company. Raised a three point two five million dollar seed in 2025 from Google, Y Combinator, Core Innovation Capital, FiDi Ventures, and strategic angels, then a twenty one million dollar Series A in February 2026 led by Insight Partners, with Y Combinator and MTech Capital participating. Reports growing annual recurring revenue fifty times over the past year and deployments across thousands of employers, with customers including carriers Reliance Matrix, New York Life, and Sun Life US, and consultants OneDigital and Daybright Financial.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Data arrives as documents, not through connectors. The extraction agents scan plan documents, contracts, proposals and census files across spreadsheets, Word documents, PDFs and other formats, infer schemas and standardize them into a single source of truth. Outbound, the platform sends email and SMS communication campaigns that link members through to a benefits microsite and decision support, and hosts the microsite and benefits hub itself. RFP intake management structures broker proposals into a navigable view. No external system is named, whether a benefits administration platform, HRIS, payroll provider, enrollment system or identity provider, and there is no integrations page, connector directory or count. No developer API, SDK or MCP surface is published.
In practice
Building a benefits plan from scattered plan documents and census files is slow manual work for your team and still has errors. Pasito's agents scan plan documents, contracts and census files, infer the schema and standardize them into a single source of truth that every plan asset is generated from.
Every enrollment season you rebuild guides, comparisons, and member communications from scratch. Pasito generates benefits guides, microsites and email and SMS campaigns from the same normalized plan data, and the campaigns link members straight into decision support.
You are a carrier or broker who needs automation but must keep your brand and stay compliant. Pasito deploys fully white labeled AI agents under your brand on a platform with HIPAA safeguards and an independently audited SOC 2 Type II report.
Sources & related URLs
Agentic Index coverage score
7.5 / 14 capabilities · 54%
| Integrations & Tool Calling | Partial |
|---|---|
|
Campaigns send over email and SMS and automatically link members through to a microsite and decision support, so the platform writes into a member's inbox and phone and hosts the destination. Document ingestion accepts "spreadsheets, Word documents, PDFs, and more". RFP intake management "structures RFP documents and data into a clear, navigable view", so the inbound path handles whatever a carrier or broker actually sends. Pasito names no benefits administration platform, carrier system, HRIS, payroll provider, enrollment system or identity provider. The site has no integrations section, and no connector directory or count is published. Carriers, brokers and employers keep their data in benefits admin systems and HRIS, but data reaches Pasito as documents and files, not through a connected system. Sourcepasito.ai/who-we-serve/employers, /who-we-serve/insurance-carriers and homepageread 2026-09-12 |
|
| Workflow Orchestration | Full |
|
One flow runs from proposal to utilization, and the sequence is Pasito's own process, not one the customer designs. Pasito says it "simplifies benefits from proposal to utilization with AI agents trained to understand plans and execute workflows at scale". Inside that span, documents are scanned and their schemas structured, assets are created, campaigns are coordinated and delivered, and members are answered year round. The steps run as one chain, not as separate tools, with "every asset standardized, created, and delivered through AI workflows for every client". Highlight sheets, enrollment presentations and engagement communications are all generated from the same normalized layer. They then go out through linked email and SMS campaigns into a microsite and decision support. Extraction feeds generation, which feeds delivery, which feeds support. Several participants take part. The carrier or broker uploads plans and reviews assets, the extraction and generation agents do the work, the employer HR team runs enrollment, and the member receives a campaign and asks the assistant a question. Each client also has an assigned success manager. One worked path runs from a carrier's plan document to an employee's coverage question being answered from it. Pasito names no orchestration runtime, describes no handoff between agents and has no authoring or workflow design surface. The customer's documents set the parameters of Pasito's benefits process, and the customer does not build the flow. Pasito does not describe any quoting or claims work. Sourcepasito.ai/what-we-do, /who-we-serve/insurance-carriers and /who-we-serve/employersread 2026-09-12 |
|
| Knowledge Grounding & RAG | Full |
|
Agents are grounded in the customer's own plan documents, which Pasito builds into a maintained, normalized store. The agentic AI "scans every document, structures new schemas, and standardizes client information for every asset", extracting "from multiple file types, including spreadsheets, Word documents, PDFs, and more". Plan documents, contracts, proposals and census files arrive unstructured and leave as a normalized store. Schema inference is the core of the product, not a preprocessing step. The normalized layer is "the same source of truth as the rest of the platform", and every downstream asset is generated from it. It persists between runs and across products instead of being assembled for each request, and it scales past any context window across a carrier's whole book of plans. Administrators configure the assistant with source documentation, so it answers a member's coverage question from the employer's actual plan, not from a model's general knowledge. Proposal comparison and RFP intake structure documents "into a clear, navigable view", so people can query the store as well as the agents. Sourcepasito.ai/who-we-serve/insurance-carriers, /what-we-do and homepageread 2026-09-12 |
|
| Human Oversight & Guardrails | Partial |
|
Admins configure the AI benefits assistant before it runs, with "admin controls for tone, custom instructions, and source documentation". An administrator sets which documents the assistant may answer from, which keeps it from answering a coverage question from anything but the plan the employer actually bought. Clear disclosures tell members when AI produced the content, and the AI governance framework adds a validation gate before deployment. The assistant "answers benefits questions 24/7" directly to employees. Pasito describes no queue, checkpoint, confidence threshold or human sign off before a member receives an answer about their coverage. Benefits guides, microsites and comparison documents are generated for a broker or carrier to review before publishing. The customer reviews those drafts, and Pasito describes no approval gate built into the product. Sourcepasito.ai homepage and /securityread 2026-09-12 |
|
| Security, Identity & Governance | Full |
|
Prescient Assurance independently audits Pasito's SOC 2 Type II, "confirming our controls for security, availability, and confidentiality". Customers can ask for the report through a Request our SOC 2 report form or a separate audit report request route. HIPAA compliance covers physical, technical and administrative safeguards for protected health information, Business Associate Agreements where applicable, and secure handling under federal regulations. Access runs under a zero trust model. Pasito says "we implement role-based access controls and least privilege principles, with routine reviews and revocation protocols". That RBAC, with periodic review and a revocation process, covers customer facing access. Each client's data is isolated in a dedicated trust zone, so data from different clients is never mixed. Data is encrypted with AES-256 at rest and TLS 1.2+ in transit, and the platform is monitored 24/7. Internal and third party experts run regular penetration tests, vulnerability scans and static and dynamic code analysis. All staff and contractors pass background checks and sign confidentiality agreements, and security training, including threat simulations, is continuous. Sourcepasito.ai/securityread 2026-09-12 |
|
| Observability & Auditability | Partial |
|
Administrators configure the assistant with source documentation alongside tone and custom instructions, so an answer can be traced back to the plan document it came from. Pasito commits to "clear disclosures", and "users are always informed when AI is used to generate or enhance content". Engagement analytics show employers and advisors where engagement is breaking down. Pasito names no run log, action history, audit trail, export or retention period. It does not say whether a record is kept of which documents the extraction agent read, what schema it inferred, what it changed or what the assistant told a given member. The 24/7 monitoring is Pasito's own staff watching the platform for threats, not a trail for customers. Request audit reports is a DocuSign form for the SOC 2 report. Sourcepasito.ai/who-we-serve/insurance-carriers, homepage and /securityread 2026-09-12 |
|
| Memory & State Persistence | Not documented |
|
Under Pasito's Zero Data Retention Agreement with Anthropic, "inputs and outputs are processed ephemerally". Pasito says "the data exists only for the duration of the session and is immediately discarded once the interaction is complete", with "no data stored, logged, or used to train AI models". The agents keep no memory across sessions, by design, on a product that handles protected health information. The data layer holds the customer's own records, with normalized plan documents, contracts and census files kept as a source of truth. That is the application's data model and the corpus the agents read, not agent memory. Models are "customized for each client using only their data", so learning is absorbed into a model for each tenant. That improves the system but gives the customer no store to point at. No employee data enters training without explicit approval. Sourcepasito.ai/security and homepageread 2026-09-12 |
|
| Deployment & Data Residency | Not documented |
|
No hosting region, data location or hosting provider is published, although Pasito handles protected health information under HIPAA for US carriers and employers. Each client's data lives in a dedicated trust zone, which Pasito says is "architected to prevent any cross-client access or data co-mingling". That trust zone is a boundary inside Pasito's cloud, not an environment the customer selects, controls or places. HIPAA and SOC 2 cover compliance, not where data resides. Pasito names no region list or customer environment and does not say where data is located. Sourcepasito.ai/securityread 2026-09-12 |
|
| Prebuilt Agents, Templates & Packs | Full |
|
Eight preconfigured units, each with its own page, make up a catalog arranged under three audiences in a menu. Benefits consultants get Benefits guides, Benefits microsites, the AI benefits assistant and Proposal comparison. Insurance carriers get Decision support, Marketing documents, the AI benefits assistant and RFP intake management. Employers get Decision support, Benefits microsite and benefits hub, the AI benefits assistant and Year-round communications. A customer browses that menu, picks what applies, and each unit does its work once selected. Each unit stands alone and does a different job for different people. The AI benefits assistant answers coverage questions on its own, and proposal comparison structures a broker's RFP documents into a navigable view without it. The units ship configured, not built by the customer. Pasito calls them "AI agents trained to understand plans and execute workflows at scale" and "custom AI agents and workflows", with assets "templatized" and every asset "standardized, created, and delivered through AI workflows for every client". Templates and prebuilt agents arrive working, with no authoring needed. The units are built for the benefits vertical. There is no library a customer adds to, no install path and no route for third party contributions, and the catalog covers Pasito's own product only. White label branding changes how these agents are presented. Sourcepasito.ai/what-we-do, /who-we-serve navigation and /who-we-serve/insurance-carriersread 2026-09-12 |
|
| Triggers & Channel Coverage | Full |
|
Employers can "build and launch email and SMS benefits communications campaigns that automatically link to your benefits microsite and decision support". Email and SMS are two named outbound channels. The benefits microsite and benefits hub is a third, a persistent surface members return to. The AI benefits assistant is a fourth, a conversational one that answers benefits questions 24/7. Campaigns are coordinated so a send lands a member on the microsite and in decision support, chaining the channels instead of sending four separate messages. Year-round communications is its own product line, separate from open enrollment campaigns, so the customer sets a cadence across the benefits year instead of sending a single blast. The assistant runs 24/7 with no operator needed to start it. Decision support is "proactive and personalized". It leads with a recommendation, "removing lengthy intake forms", and identifies best fit plans from preset demographic information, so the agent opens the conversation from a data condition instead of waiting to be asked. The enrollment cycle is the recurring event the whole product is built around. Only two outbound channels are named. Pasito names no voice channel, no chat platform such as Teams or Slack, and no WhatsApp. Sourcepasito.ai/who-we-serve/employers, /employers/decision-support and /employers/year-round-communicationsread 2026-09-12 |
|
| Model Flexibility & Routing | Not documented |
|
Anthropic, the AI provider behind Claude, is the one model provider Pasito names, and Pasito maintains a Zero Data Retention Agreement with it. Customers get no choice of model. Pasito publishes no model picker, admin setting, per agent assignment or bring your own key path. The admin controls cover tone, custom instructions and source documentation, which shape what the agent says and reads, not the model that runs it. Pasito says "Models are customized for each client using only their data and Pasito's generalized training data, never shared or repurposed". That tunes behavior to each customer but does not let the customer choose a provider, and the plural refers to those tuned variants. Strict consent policies decide whether employee data enters training at all. Pasito chooses the model itself. Sourcepasito.ai/security and homepageread 2026-09-12 |
|
| APIs, SDKs & MCP Extensibility | Not documented |
|
No API reference, endpoint list, authentication guide, SDK, webhook or MCP server is published, and the site has no developer, docs, API or integrations section. app.pasito.ai is the application itself, behind a login, and there is no docs or dev subdomain, llms.txt or page for assistants. White labeling deploys an agent under a client's own brand, with its logo, tone and domain. That changes presentation only. Pasito ships no way for clients to build their own agents in the workspace. Pasito sells through demos to carriers, brokers and employers and assigns a success manager to every client. There is no self serve tier a developer could reach. Sourcepasito.ai/security, homepage navigation and footerread 2026-09-12 |
|
| Testing, Debugging & Optimization | Full |
|
An AI governance framework tests changes before launch and evaluates them after. Under Pasito's generative AI use policy, "pre-deployment testing validates AI output accuracy, data integrity, and ethical alignment before launch". A change is tested before release, and accuracy is what gets measured. After launch, "ongoing evaluations ensure AI tools meet performance, fairness, and safety standards". This is a controlled loop against stated standards, not a live dashboard. An incident response protocol closes it, with the data team investigating and resolving deviations. A feedback loop lets clients and users submit concerns directly, and these feed into platform updates and enhancements. Pasito publishes no scoring scale, test set construction, pass threshold or verdict format, so customers can see that a validation occurred but not its result. The test harness is Pasito's own and gates Pasito's releases. Pasito describes no sandbox where customers test a changed agent before it reaches members. On bias, Pasito says "we follow industry AI ethics guidelines", a commitment to guidelines, not an audit, and fairness is among the standards evaluated after launch. Sourcepasito.ai/securityread 2026-09-12 |
|
| Browser & Computer Use | Not documented |
|
The work is document processing and content delivery, and neither touches an interface. Pasito's agentic AI "scans every document, structures new schemas, and standardizes client information", extracting from spreadsheets, Word documents and PDFs. Assets are generated from that store, campaigns go out over email and SMS, and members read a microsite and ask a conversational assistant. Every step is file parsing, generation or messaging. Pasito describes no hosted or local browser, desktop session, remote or local computer control, RPA or browser extension. Its inputs arrive as documents a client uploads, not as pages someone browses. When a PDF's layout changes, that is a parsing problem, not a UI automation one, and Pasito handles it by inferring schemas, not by clicking. The benefits microsites and benefits hub are surfaces Pasito builds and hosts for members to visit, not someone else's interface that it operates. White labeled deployment of agents under a client's brand changes only their presentation. Sourcepasito.ai/what-we-do, /who-we-serve/insurance-carriers and homepageread 2026-09-12 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Pricing
Quote only
Not published. No unit, tier or minimum appears anywhere. The only routes are Request a demo and a login for existing clients.
What is public
Pasito publishes nothing on price, with no pricing page and no rate card. The site navigation runs What we do, Who we serve, Resources and Who we are alongside a Login for existing clients and a Request a demo call to action, and every commercial route on the homepage, the audience pages and the security page resolves to the demo form. What is published in place of price is the shape of the offering, eight named product units across three audiences. Benefits consultants get benefits guides, benefits microsites, an AI benefits assistant and proposal comparison. Insurance carriers get decision support, marketing documents, an AI benefits assistant and RFP intake management. Employers get decision support, a benefits microsite and hub, an AI benefits assistant and year round communications. A client success manager is assigned to every client for onboarding, training and ongoing support. The SOC 2 Type II report is obtainable through a published request form, and Terms of Service, a Privacy Policy and a Security page are published.
Billing mechanics
Quote only, through a demo request, with no billing unit published. Whether billing runs per employer group, per covered member, per product module or as a flat platform fee is not stated.
Cost watchouts
Expanding across product lines such as health, life and retirement, or across more employer groups, widens what is licensed, and no tier or price for that expansion is published. Whether forward deployed implementation is billed separately is not published either.
Variable cost rationale
No metering is published and no billing unit is stated. The activity that could drive cost is member facing. The AI benefits assistant answers members 24/7, and campaigns go out over email and SMS to whole employee populations, so use tracks covered lives rather than admin seats. For a carrier or consultancy it also tracks a book of clients that grows. Nothing published caps the bill. The platform is sold through demos with an assigned client success manager, which points to a committed annual fee rather than per message billing, and document extraction is bounded by the plan year rather than continuous.
Additional watchouts
Pasito sells three different product sets under one platform, so a carrier, a broker and an employer are buying different things. For a carrier or consultancy the value scales across a whole book of clients, and nothing published says whether the license covers the employers served or whether each is billed, or what happens as that book grows during the term. Deploying agents under the client's own brand is published as a white labeling capability with no price attached. The SOC 2 Type II report, relevant to a HIPAA review, is obtainable through a published request form, while no data residency is published at all, which matters on a product handling protected health information. There is no trial and no self serve tier, and every route is a demo.
Sales call required
Yes, required for paid access
Free / trial
None published
Lowest paid plan
Not published
Key ambiguities
No pricing unit is published, so whether cost tracks the carriers, brokers, employer groups or members served is unknown. The live question is which side of the chain pays and for what. Pasito sells to three distinct audiences with different products (benefits consultants, insurance carriers and employers), and a carrier templatizing highlight sheets across its whole book is a different commercial shape from an employer running one enrollment. Nothing published says whether a carrier's license covers the employers it serves, or whether those employers are billed separately. White labeled deployment under a client's own brand is a stated capability with no stated price, and on a product sold to carriers who resell to brokers that is likely a material line. No trial is published. The only routes are a demo request and a client login.
Missing data
Everything numeric, and the unit with it. There is no pricing page in the navigation, which runs What we do, Who we serve, Resources, Who we are, Login and Request a demo, and no rate, tier, minimum, contract length or billing cadence appears anywhere. Also unpublished are the billing basis (per employer group, per covered member, per product module or a flat platform fee) and whether the eight named product units are licensed together or separately. Nor is it published what white labeled deployment costs, whether a carrier license extends to the brokers and employers downstream of it, or whether the assigned client success manager is bundled or chargeable.
Related vendors
- 11th Estate — Agentic platform whose AI engine scans markets, matches an…
- Abacum — FP&A platform for growing companies whose AI Modeler builds, edits…
- Adopt AI — AI native accounting and bookkeeping firm whose agents run…
- Aera Technology — Decision intelligence platform where an always on agent executes…
- Akro AI — On premise operational intelligence that automates document heavy…
- alfred_ — AI executive assistant that triages the inbox overnight and scores…
Alternatives to Pasito
The closest documented capability profiles to Pasito among enterprise operations agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Firmbase5.5 / 14A lighter documented profile than Pasito
- ketteQ9.5 / 14Adds documented APIs, SDKs & MCP Extensibility
- Logility6.5 / 14Fuller documented coverage on Human Oversight & Guardrails
- o9 Solutions6.5 / 14Adds documented Model Flexibility & Routing
- Obin AI9.5 / 14Adds documented Deployment & Data Residency
- Phaidra7.5 / 14Fuller documented coverage on Integrations & Tool Calling and Human Oversight & Guardrails
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded