Comic Inventory AI
We built an AI-powered SaaS platform that transforms physical comic books into structured, searchable, and valued digital assets. Using Softr as the Access Layer, the platform connects to a complex backend powered by Airtable, Make, OpenAI Vision, Uploadcare, and Stripe. Users upload a photo of a comic book. AI extracts metadata, estimates grade, assigns confidence scoring, and routes records through a structured Intake → Canonical promotion workflow. Softr powers the user dashboards, review queues, filtering system, role-based visibility, subscription gating, and account-level data isolation. This project demonstrates deep Softr integration with multi-table relational architecture, advanced permission logic, AI-driven workflows, automation orchestration, and subscription-based SaaS scalability.
From the partner
This project is an AI-powered comic collection SaaS platform that converts physical comic books into trusted, structured digital assets.
The core business objective:
- Reduce friction in logging collectibles
- Eliminate grading and valuation ambiguity
- Turn private collections into searchable, scalable assets
- Create a SaaS platform capable of expanding from individual collectors to comic stores
As outlined in the platform architecture, the system operates on a three-layer model:
- Capture Layer – Image ingestion and AI extraction
- Truth Layer – Canonical data normalization and validation
- Access Layer – User interaction (powered by Softr)
Softr is the primary interface layer that makes this entire system usable.
Technical Architecture
Backend Data Infrastructure
Database: Airtable
Structured into protected layers:
- Intake (raw AI-generated records)
- Comics (canonical verified assets)
- Users
- Publishers
- Series
- Alias tables for normalization
- Subscription and credit tracking
Strict separation ensures: - No unverified AI data reaches canonical records
- Scalable SaaS-ready schema
- Multi-user data isolation
Automation & AI Layer
Make (Integromat) orchestrates:
- Webhook-triggered ingestion
- OpenAI Vision image analysis
- Structured JSON extraction
- Confidence scoring
- Conditional routing (Auto-Approve vs Needs Review)
- Duplicate detection
- Canonical promotion rules
- Subscription enforcement
- Stripe billing events
- Uploadcare image transformation
This includes: - Multi-scenario orchestration
- Error handling routes
- Status-based triggers
- Conditional branching based on plan level
- Credit usage optimization per comic processed
AI Agents
The platform does not use a single AI call — it uses a modular agent design:
- Intake Vision Agent
- Matching/Normalization Agent
- QA/Duplicate Detection Agent
- Valuation Agent (planned expansion)
- Query Agent for chat-based exploration
Each agent produces structured output and does not override canonical truth without validation.
Softr’s Role (Access Layer)
Softr powers the entire user experience:
- User Dashboard
- Authenticated login
- User-specific data filtering
- Subscription-tier gated features
- Upload tracking and usage monitoring
Intake Review Queue - Dynamic filtering of “Needs Review” records
- Editable metadata fields
- Promotion actions
- Confidence visibility
- Admin-level override capabilities
Comic Detail Pages - Relational record linking
- Dynamic field population
- High-resolution image rendering via Uploadcare
- Tag filtering
- Grade and valuation display
Collection Views - Advanced filtering (publisher, grade, tags, value)
- Conditional display logic
- Searchable lists
- Responsive design for mobile and desktop collectors
Security & Data Isolation - User-level record visibility
- Role-based permissions
- Owner-based filtering logic
- Admin-only moderation views
- Plan-based feature gating
Business Model Integration
Softr integrates directly with:
- Stripe subscriptions
- Plan-based upload limits
- Credit tracking
- Tiered storage access
- Feature gating (bulk uploads, advanced analytics, valuation)
This required: - Dynamic field rules
- Conditional visibility
- Usage enforcement logic
- Subscription event handling via Make → Airtable → Softr
Complexity Highlights
This is not a simple CRUD front-end.
It includes:
- AI-driven data extraction with confidence scoring
- Structured intake → canonical promotion pipeline
- Multi-layer data validation
- Alias-based normalization to prevent misidentification
- Duplicate detection logic
- Tiered SaaS architecture
- Image transformation pipelines
- Automation cost modeling
- Scalable schema ready for Postgres migration
- Multi-agent AI design
Softr serves as the clean, intuitive access point to a deeply engineered automation and intelligence system.
Why This Project Matters for Softr
This implementation demonstrates:
- Enterprise-level relational database design
- Advanced automation orchestration
- AI integration beyond simple API calls
- Subscription-based SaaS modeling
- Clean separation between raw AI data and trusted canonical data
- Scalable architecture from MVP to full SaaS expansion
It shows how Softr can sit on top of a highly complex backend and deliver a polished, secure, and commercially viable SaaS product.
Demo
Gallery images
Data sources
Partner details
The e-four Corporation
efour is a strategic growth and systems consultancy based in Elgin, Illinois, specializing in building platforms with Softr, MAKE and Airtable. We help founders and SMB leadership teams transform disconnected tools into scalable client portals, subscription platforms, websites and operational systems that increase efficiency and enterprise value. Our approach blends executive-level marketing strategy with hands-on automation expertise, so what we build isn’t just functional, it’s aligned to measurable business outcomes.