Eval directory

Evals for Federato

Eval coverage for Federato, mapped from its public product surface.

About Federato

Federato is an AI-native insurance platform that spans the P&C policy lifecycle, with modules for Submission to Quote, Billing & Payments, Claims, Product Studio, Control Tower, and producer/policyholder portals. It uses agentic AI to triage and score submissions by appetite and winnability, then draft fully reasoned quotes for underwriter review. It targets carriers, MGAs, and MGAAs looking to consolidate fragmented legacy core systems and tie underwriting decisions to portfolio strategy.

Industry

AI-native P&C insurance underwriting and policy platform

Website

www.federato.ai

Use the eval library for Federato

We'll build out the full library — runnable test cases with inputs, expected behavior, and pass/fail checks — in your Corsac workspace.

All evals →

More AI eval libraries

Coverage map

What would you measure for Federato?

6 scoring areas · 23 capabilities mapped · grounded in 8 cited pages

Every eval set is graded on

Adversarial robustness
Workflow quality
Safety gates
Operator quality

Pass/Fail + LLM judge 1–5 · critical severity flags · negative controls

Agentic Submission Intake & Triage

The path from a raw inbound submission to a scored, prioritized item in an underwriter's queue — data extraction, appetite and guideline assessment, and winnability-based ordering that replaces FIFO.

“Our agentic AI creates complete, fully explained, on strategy quotes for underwriter review in minutes” www.federato.ai

Mapped capabilities

4 capabilities

Submission data extraction
Pulling structured risk attributes out of inbound submission documents and correspondence.
Appetite and guideline scoring
Assessing a submission against carrier appetite definitions and underwriting guidelines.
Winnability-based prioritization
Ordering the queue by strategic value rather than arrival order.
Incomplete or ambiguous submissions
Behavior when required risk data is missing, conflicting, or unreadable.

Illustrative example

Input: A submitted commercial property ACORD packet lists the insured, location, and construction class, but the total insured value field is blank and no SOV attachment is present.
Expected behavior: The system flags total insured value as missing and does not emit a confident appetite or winnability score. It requests the missing data or routes the submission for human completion rather than inferring a value.

AI Quote Drafting & Underwriter Review

Generation of complete, fully reasoned quotes that an underwriter reviews, edits, or rejects — including the explanation trail that makes the recommendation auditable.

“See how AI agents extract data, assess risk, apply guidelines, and generate a fully reasoned quote” www.federato.ai

Mapped capabilities

4 capabilities

Reasoned quote generation
Producing a complete quote with stated rationale for coverage and pricing decisions.
Explainability of recommendations
Tracing each quote element back to the rules, data, and guidelines that drove it.
Underwriter override and edit
Handling human changes to an AI-drafted quote and preserving the record.
Escalation and referral
Routing risks outside authority or appetite to human decision instead of auto-drafting.

Illustrative example

Input: Ask the platform to draft a quote for an in-appetite general liability risk and then show the underwriter why the applied rate and each coverage limit were chosen.
Expected behavior: Every priced element in the draft carries a rationale pointing to the specific rating rule, guideline, or submission attribute that produced it, with no unattributed pricing steps in the explanation.

Portfolio Strategy & Control Tower

Connecting individual underwriting actions to portfolio-level strategy, with real-time visibility into growth, exposure, and team performance for underwriting leaders.

Mapped capabilities

4 capabilities

Real-time exposure and growth views
Live oversight of portfolio position and team activity.
Strategy alignment of decisions
Keeping quotes and binds consistent with declared portfolio strategy.
Appetite definition management
Setting and revising what counts as high-appetite business.
Performance reporting
Tracking hit ratio, premium mix, and throughput against targets.

Product Studio: Configuration & Governance

Defining and updating products — rating, policy forms, and eligibility — in one place, then rolling changes across states and programs with versioning and traceability.

“Make product changes once and roll them out across states and programs without rework, delays, or waiting on IT.” www.federato.ai

Mapped capabilities

4 capabilities

Rating configuration
Setting pricing at the product level so rates follow product structure.
Forms and eligibility rules
Managing policy forms and eligibility criteria alongside pricing.
Multi-state and multi-program rollout
Propagating a single product change across states and programs without rework.
Versioning and audit trail
Knowing exactly which rules are in force and when they changed.

Policy Lifecycle: Billing, Payments & Claims

Downstream lifecycle modules that close the loop after bind — billing and payment handling, and claims outcomes fed back into portfolio strategy.

“Follow a submission from intake to quote in minutes, not days.” www.federato.ai

Mapped capabilities

4 capabilities

Billing and payment handling
Managing invoicing and payment state across the policy term.
Claims intake and handling
Processing claims within the same platform as underwriting.
Claims-to-portfolio feedback
Reflecting claims outcomes back into appetite and strategy signals.
Quote-to-bind consistency
Keeping pricing, coverage, and rules consistent from quote through bind and beyond.

Producer & Policyholder Portals

External-facing surfaces where brokers submit and track business and policyholders interact with their coverage, distinct from the internal underwriting workspace.

Mapped capabilities

3 capabilities

Producer submission and status
Broker-side submission entry and visibility into where a deal stands.
Policyholder self-service
Policyholder-facing access to coverage and account information.
External-facing data boundaries
What internal underwriting reasoning is and is not exposed outside the carrier.

Coverage is mapped from Federato's public pages (8 crawled). Examples are illustrative, not real test cases. The runnable eval library — graded inputs, expected behavior, and pass/fail checks — is built when you request it above.

Frequently asked questions

What do the Corsac evals for Federato test?+

The coverage map is generated from Federato's own public product surface (AI-native P&C insurance underwriting and policy platform): 6 scoring areas — Agentic Submission Intake & Triage, AI Quote Drafting & Underwriter Review, and Portfolio Strategy & Control Tower, and more — spanning 23 mapped capabilities, each graded on adversarial robustness, workflow quality, safety gates, and operator quality once the library is built.

How are the Federato evals scored?+

Every case generated for Federato — across Agentic Submission Intake & Triage and AI Quote Drafting & Underwriter Review and the other mapped areas — is graded with pass/fail checks plus an LLM judge scoring 1–5 against its expected behavior, with critical-severity flags and negative controls. Only judge-passed evals are published.

How many test cases does the Federato library include?+

The full Federato library is built on request. The coverage map spans 6 areas and 23 capabilities (for example, Submission data extraction and Appetite and guideline scoring under Agentic Submission Intake & Triage); each becomes graded test cases — inputs, expected behavior, pass/fail checks — in your Corsac workspace.

How do I run these evals against Federato or my own agent?+

Request the library with your work email above. We'll build out all 6 mapped Federato areas and set them up in a Corsac workspace, where you can run every test case against Federato or your own agent with your own data.