• No instant quoting
  Customers had to request quotes manually, creating delays and limiting throughput.
• System fragmentation
  ERP and CRM systems were not integrated in a way that allowed real-time pricing or order creation.
• No end-to-end digital flow
  There was no unified journey from CAD upload → validation → pricing → order confirmation.
• Complex, non-transparent pricing
  Pricing depended on multiple variables:

• • machining time
  • material usage efficiency (nesting)
  • customer-specific contracts and discounts

These calculations were done manually or semi-manually, increasing the risk of errors and inconsistencies.

• Scalability limitations
• The existing workflow could not handle increased demand without proportionally increasing operational effort.

Solution:

We designed and built ePlast Editor Core as a modular, service-oriented platform that automates the entire quoting and ordering lifecycle.

Key capabilities:

• Self-service CAD upload and validation
  Users upload STEP/DXF/STL files, which are automatically validated via background workers.
• Interactive 3D configurator
  A browser-based interface allows users to visualize and configure parts before ordering.

• Real-time pricing engine
  Pricing is dynamically calculated based on:
  • machining time estimation
  • nesting efficiency
  • contract and campaign rules

• Platform orchestrator
  A state-machine-based orchestration layer coordinates multiple services (timing, nesting, pricing), ensuring reliability and consistency across flows.

• Automated order pipeline
  Once confirmed, orders are processed and synced directly with the ERP system.

Core automated flows:

1. Price Update — instant recalculation on configuration changes
2. Order Submission — quote → validated order → ERP sync
3. Cart Update — external changes trigger full recalculation
4. CAD Validation — async validation via distributed workers
5. Material Filtering — dynamic availability based on part specs
6. Nesting Jobs — optimized material usage across grouped parts
7. The result is a system that behaves like a real-time pricing engine rather than a static quoting tool.

Technologies:

We selected technologies based on performance, scalability, and the ability to handle computationally intensive geometry processing.

Backend

• Python 3.11 + FastAPI — high-performance async APIs with automatic OpenAPI generation
• SQLAlchemy / SQLModel — type-safe ORM for structured data handling
• PostgreSQL 17 — robust relational database for complex domain models
• Redis + Arq — async task queue for long-running jobs
• Alembic — database migration management
• Geometry & CAD Processing
• FreeCAD / CadQuery-OCP — parsing and processing STEP/DXF files
• Trimesh + Shapely — 3D mesh analysis and 2D polygon operations for nesting
• Cloud & Storage
• Azure Blob Storage — CAD file storage
• Azure Storage Queue + Azure Batch — distributed validation workers

Frontend

• React + TypeScript — scalable, type-safe UI
• Vite + SWC — fast build tooling
• TanStack Router/Query — state and data synchronization
• Tailwind CSS + shadcn/ui — consistent UI system
• Three.js — real-time 3D rendering in the browser
  

Infrastructure & DevOps

• Docker + Docker Compose — environment consistency  
• Traefik + Nginx — routing and static content delivery  
• GitHub Actions — CI/CD automation  
• Sentry — monitoring and error tracking

This stack allowed us to balance real-time responsiveness with heavy asynchronous computation.

Results:

We built ePlast Editor Core — a full-stack platform that turns a manual quoting process into a self-service, automated online ordering experience for CNC-cut custom parts.

• Fully automated quoting process
  What previously required manual effort is now completed in seconds.

• Reduced operational workload
  Engineers are no longer involved in routine quoting, freeing them for higher-value tasks.

• Improved customer experience
  Customers can upload files, get instant pricing, and place orders without back-and-forth communication.

• Accurate and consistent pricing
  Centralized pricing logic eliminates discrepancies and reduces human error.

• Scalable architecture
  The system supports increasing demand without proportional growth in operational cost.

• Seamless system integration
  ERP and pricing systems are now part of a unified, real-time workflow.

• Production-ready platform delivered in 6 months
  A team of 5 engineers successfully designed and implemented the system from scratch, including complex geometry processing and multi-service orchestration.