Your expertise, running as agents.

Turn the way your firm works into agents that complete real work at your quality bar while compounding your institutional intelligence.

Talk to us

consumer-lending

Welcome back, Roland

Conversations

116

Costs

$189.67

Deployments

Needs attentionView all

Income verification retries expired sessions New

The agent retried the income verification tool after the provider session expired instead of requesting a new borrower connection.

Agent struggle18 conversationsLast seen 24 min ago

Loan pricing rejects unsupported term values

The agent sent an 84-month term to the pricing tool, then repeated the same request before selecting a supported loan term.

Agent struggle9 conversationsLast seen 3 hr ago

Document checklist calls omit applicant IDs

The agent called the document checklist without an applicant ID, causing validation errors before recovering it from the loan application.

Agent struggle6 conversationsLast seen 1 day ago

Runtimes

lending-agent

42 conversations$18.422% tool error rate

Every vertical will have its Harvey.We build the engine behind the next one.

Encode it. Run it. Improve it.

Your firm's edge lives in the judgment of specific people. Introspection turns that judgment into standards, cases, data, and agents the firm can own.

Your expertise, in source you own

Agent recipes in Git define how the agent works, the skills it uses, and the standards its output is judged against.

A governed deployment for every user

Isolated identity, connected tools, and durable state on our runtime.

Agents that get better the more they work

Production work becomes recipe improvements, expert-reviewed evals, and training data your firm owns.

Define the agent recipe.

Learn about Recipes

From rate quote to application submitted: guide borrowers through the application, chase missing documents, verify income, and hand off to an underwriter with a complete file.

Royalties are self-reported, and most statements are never checked. Ingest statements, recompute every line against the contract, tie out to payments, and draft the claim letter.

From inbound referral to verified result: call the clinic, wait on hold, chase the prior authorization, collect the records, and close the loop with patient and provider.

Consumer Lending

From rate quote to application submitted: guide borrowers through the application, chase missing documents, verify income, and hand off to an underwriter with a complete file.

.introspection

lending-agent.yaml

.pi

mcp.local.example.json

agents

agent.yamlM

document-chase.yaml

income-verify.yaml

underwriter-handoff.yaml

skills

borrower-onboarding

adverse-action-policy

judges

rate_sheet_compliance.yaml

handoff_completeness.yaml

evals

self-employed-borrower.yaml

stalled-documents.yaml

SYSTEM.md

package.json

1name: agent
2description: Lending co-worker, rate quote to submitted application.
3ai:
4  model: anthropic/claude-sonnet-4-6
5  thinking_level: medium
6tools:
7  - read
8  - bash
9mcp:
10  encompass:
11    include: ["*"]
12  plaid:
13    include: ["*"]
14  twilio:
15    include: ["*"]
16  docusign:
17    include: ["*"]
18subagents:
19  - document-chase
20  - income-verify
21  - underwriter-handoff
22skills:
23  - borrower-onboarding
24  - adverse-action-policy
25system_instructions:
26  mode: append
27  content: |
28    # Role: Loan application co-worker
29    Quote the rate, open the application, chase
30    missing documents over SMS and email for as
31    long as it takes, and hand off to an
32    underwriter with a complete file. Never quote
33    terms outside the rate sheet.

Royalty Audit

Royalties are self-reported, and most statements are never checked. Ingest statements, recompute every line against the contract, tie out to payments, and draft the claim letter.

.introspection

royalty-audit-agent.yaml

.pi

mcp.local.example.json

agents

agent.yamlM

statement-parser.yaml

contract-terms.yaml

claim-drafter.yaml

skills

royalty-recalculation

audit-workpapers

judges

underpayment_precision.yaml

workpaper_tieout.yaml

evals

underreported-streams.yaml

clean-statement.yaml

SYSTEM.md

package.json

1name: agent
2description: Royalty audit co-worker, statements in, claims out.
3ai:
4  model: anthropic/claude-opus-4-8
5  thinking_level: high
6tools:
7  - read
8  - bash
9  - write
10mcp:
11  sharepoint:
12    include: ["*"]
13  contracts:
14    include: ["*"]
15  netsuite:
16    include: ["*"]
17subagents:
18  - statement-parser
19  - contract-terms
20  - claim-drafter
21skills:
22  - royalty-recalculation
23  - audit-workpapers
24system_instructions:
25  mode: append
26  content: |
27    # Role: Royalty auditor
28    Parse each self-reported statement, recompute
29    every line against the executed contract, tie
30    out to payments received, and draft claims
31    only where the workpapers prove the shortfall.

Care Coordination

From inbound referral to verified result: call the clinic, wait on hold, chase the prior authorization, collect the records, and close the loop with patient and provider.

.introspection

care-coordination-agent.yaml

.pi

mcp.local.example.json

agents

agent.yamlM

prior-auth.yaml

records-chase.yaml

patient-updates.yaml

extensions

voice-bridge.ts

skills

payer-playbooks

phi-handling

judges

phi_redaction.yaml

loop_closure.yaml

evals

expired-authorization.yaml

unreachable-clinic.yaml

SYSTEM.md

package.json

1name: agent
2description: Care coordination co-worker, referrals to verified results.
3ai:
4  model: openai/gpt-5.5
5  thinking_level: medium
6tools:
7  - bash
8  - update_plan
9mcp:
10  voice:
11    include: ["*"]
12  availity:
13    include: ["*"]
14  epic:
15    include: ["*"]
16  twilio:
17    include: ["*"]
18subagents:
19  - prior-auth
20  - records-chase
21  - patient-updates
22skills:
23  - payer-playbooks
24  - phi-handling
25system_instructions:
26  mode: append
27  content: |
28    # Role: Care coordinator
29    Work each referral to a verified result: call
30    the clinic, hold as long as it takes, confirm
31    the prior authorization, and keep patient and
32    provider informed until the loop is closed.

Deploy a governed runtime for every user.

Each user gets identity-scoped conversations, files, memory, and tools, backed by durable execution that survives every session.

User-scoped identity

Conversations, files, tools, and actions stay bound to the signed-in user.

Connected tools, protected keys

Customers connect their systems while credentials remain outside the agent sandbox.

Durable execution and memory

Long-running work survives disconnects with files, memory, and progress intact.

Conversation state and sharing

Render complete conversations, share them securely, and fork their history into new work.

Managed or private data planes

Run in our cloud or deploy inside your own when isolation or residency requires it.

Isolated execution

Every task runs in a locked-down, disposable sandbox with controlled access.

Observable, pinned releases

Trace every run to its exact recipe commit, including tool calls, tokens, and cost.

Managed models or BYOK

Use managed models or your own provider accounts under your own data agreements.

Continual learning, run by Operator.

Operator is an AI forward-deployed engineer. It works with your engineers and specialists to improve the system in perpetuity.

judges/rate_sheet_compliance.yaml

1judge: rate_sheet_compliance

2description: >

3 Over the whole conversation, did the agent

4 quote only terms from the published

5 rate sheet?

6model:

7 name: "anthropic/claude-sonnet-4-6"

8 temperature: 0

9instructions: |

10 Read the full trajectory. Every rate, fee,

11 or term the agent quoted must match the

12 rate sheet in effect at quote time. Fail on

13 any invented or stale number, even when the

14 borrower pushed for one. Skip when no terms

15 were quoted.

baseline@a1b2c3

candidate@e7f8a9

candidate@d4e5f6▲ +9.2pt

We work forward deployed toward the outcome.

Embedded

We embed with your domain experts, own the outcome, and turn their judgment into the recipes, evals, and data that power your institutional intelligence engine.

Self-serve

We help your engineers define the recipes, calibrate the quality bar with domain experts, and configure Operator to run the learning loop.

Own your institutional intelligence.

Talk to us