Inspex — AI Digital Asset Evaluation & Marketplace
Client: Inspex (Founder)
AI-powered trust and valuation layer for digital-asset deals. Inspex turns verified revenue, traffic, code, and compliance evidence into defensible valuations and underwriting-grade reports — self-serve for asset owners, diligence-ready for lenders and investors — backed by 600+ evidence connectors, 38 specialist AI agents, and the built-in CompTrail compliance engine.
Context & Objective
Buying or lending against a digital business — a domain, SaaS app, ecommerce store, or content site — still runs on self-reported spreadsheets and screenshots no underwriter can trust. Inspex was built to replace that with an evidence-based system of record: connect the asset's real revenue, traffic, and code, let a fleet of specialist AI agents verify and score it, and produce a signed, hash-verifiable report a credit committee or acquirer can actually rely on.
Goals
Give asset owners an instant, self-serve estimate and a path to a defensible number; give lenders and investors an underwriting-grade diligence package with a verifiable evidence trail; and make trust and compliance a first-class part of the deal through a built-in SOC 2 / SOC 1 readiness layer and an NDA-gated deal room.
Approach
An evidence pipeline, not a questionnaire
Every evaluation begins by connecting the asset's real accounts. Inspex pulls first-party evidence through 600+ read-only connectors, normalizes it, and runs a fleet of specialist AI agents that each score one dimension of the business. A score-fusion layer combines them into a single, defensible trust and valuation result — with every number traceable back to the evidence that produced it.
38 specialist agents
Rather than one model guessing, Inspex runs purpose-built agents for revenue and traffic verification, code quality and security, unit and owner economics, customer concentration, key-person and legal/IP risk, tax compliance, market ceiling, and transfer complexity — orchestrated and reconciled into one report by a dedicated orchestrator and score-fusion stage.
CompTrail compliance layer
CompTrail extends the same evidence engine into enterprise trust: SOC 2 / SOC 1 readiness estimation, a control catalog with automated checks, drift detection, an evidence manifest, risk scoring, policy templates, and a public trust center — turning compliance from a deal blocker into a selling point.
Deal Room and deliverables
An NDA-gated deal room connects asset owners with lenders and investors. From one evidence base Inspex produces three underwriting-grade deliverables — a lender credit memo, a VC diligence pack, and an owner valuation — each signed and hash-verifiable. Valuations use comparable triangulation and revenue-multiple modeling with 10/90 confidence bands validated by backtesting.
Outcome
A production, multi-tenant platform: a Next.js 16 dashboard, a FastAPI evaluation backend, and a Celery evidence pipeline that runs 38 specialist agents over 600+ connectors. Owners get instant estimates and shareable trust scores; lenders and investors get signed credit-memo, VC-diligence, and owner-valuation deliverables from a single evidence base; and CompTrail extends the same engine into enterprise SOC 2 / SOC 1 readiness and continuous control monitoring.
Key Highlights
- 613 evidence connectors spanning revenue, banking, analytics, ads, app stores, ecommerce, SEO, CRM and support
- 38 specialist AI agents — revenue and traffic verification, code quality, unit economics, key-person and legal/IP risk, market ceiling, transfer complexity — fused into one trust score
- Signed, hash-verifiable reports built for the credit file: one evidence base, three deliverables (lender credit memo, VC diligence, owner valuation)
- CompTrail compliance engine: SOC 2 / SOC 1 / ISO 27001 / HIPAA / PCI / GDPR readiness, control catalog, automated checks, drift detection and a public trust center
- Deal Room: an NDA-gated data room connecting asset owners with lenders and investors
- Valuation methodology: comparable triangulation and revenue-multiple modeling with 10/90 confidence bands validated by backtesting
- Read-only, borrower-permissioned data access — evidence is pulled, never self-reported
Integrations
- Stripe, PayPal & BraintreeVerified revenue and processor evidence
- Plaid & bankingCash-flow and balance evidence
- Google Analytics 4 & Search ConsoleFirst-party traffic verification
- App Store Connect & Google PlayMobile app revenue and health
- Shopify & Amazon SellerEcommerce revenue and orders
- Ahrefs, SEMrush & MozDomain authority and SEO signals
- GitHub & GitLabCode quality and security evidence
- QuickBooks & XeroAccounting evidence
- OpenAILLM provider abstraction powering the agents
- 600+ connectors totalOne catalog across payments, ads, analytics, app stores and more