Miji 弥济 logo
Use cases

Operations Miji can run.

Miji is an enterprise decision automation platform, and it fits any process where work arrives, is checked against a policy, needs a decision, sometimes needs a person, and has to be explained later. Lending ships pre-built today; the others are what the same engine builds. Each card links to the platform pages that explain how it is built.

LendingShips pre-built

Manufactured-home prequalification

A retailer's application taken to a decision: credit, the maths, your chattel or land-home grid and overlays, and underwriting on the files that need it — with a record of which policy decided.

Shape

  1. 1TriggerApplication from a retailer
  2. 2ConnectorCredit pull
  3. 3Lookup tableLTV grid by home condition
  4. 4RuleEligibility policy
  5. 5Human approvalUnderwriting, if referred
InsuranceBuilt on the engine

Claims triage

Check cover, summarise the adjuster's notes, fast-track the small clear claims and route the rest to a senior adjuster.

Shape

  1. 1TriggerClaim submitted
  2. 2ConnectorPolicy lookup
  3. 3AI stepSummarise notes
  4. 4RuleFast-track eligibility
  5. 5Human approvalSenior adjuster
Healthcare adminBuilt on the engine

Prior authorisation

Match a request against coverage criteria, approve what the criteria clearly cover, and send the rest to a clinical reviewer.

Shape

  1. 1TriggerRequest received
  2. 2Lookup tableCoverage criteria
  3. 3RuleAuto-approval policy
  4. 4Human approvalClinical review
  5. 5ConnectorNotify the provider
OperationsBuilt on the engine

Vendor and customer onboarding

Collect the documents, check them against your policy, and get sign-off from every team that has to give it before the account is created.

Shape

  1. 1TriggerForm submitted
  2. 2FileDocuments held against the case
  3. 3RuleOnboarding policy
  4. 4Human approvalFinance and legal — all
  5. 5ConnectorCreate in your ERP
Legal & complianceBuilt on the engine

Regulatory change review

Route new rules and guidance through review, flag what affects you, and escalate to counsel before the deadline.

Shape

  1. 1TriggerNew item published
  2. 2AI stepSummarise and classify
  3. 3RuleAffects us?
  4. 4Work queueCompliance review
  5. 5Human approvalCounsel sign-off
FinanceBuilt on the engine

Invoice exceptions

Match invoices against purchase orders and send only the mismatches to someone who can approve them.

Shape

  1. 1TriggerInvoice received
  2. 2ConnectorFetch the purchase order
  3. 3CalculationMatch and variance
  4. 4RuleWithin tolerance?
  5. 5Human approvalAP approver

The same four things, whatever the operation

Described, then reviewed

You describe the operation and Miji's AI drafts the workflow. You review it and change it before anything runs.

How it works

Deterministic where it counts

Rules, grids and calculations give the same answer for the same inputs. AI steps sit inside that, where judgment helps.

How it works

A person where it matters

Approval gates route by threshold and resolve by any-one, all or majority. Everything else runs without waiting.

How it works

A record that holds up

Every action, decision and approval is recorded against the run on a hash-chained audit trail.

How it works

Have a different operation in mind?

Bring it to a demo. Tell us what arrives, what decides it and who has to sign off, and we'll show you how it would run.