Quotation and fulfilment platform · The AM Lab

Digital Manufacturing Platform

Quoting a manufactured part is an expert reading a file. The platform keeps the expertise and removes the waiting.

Built for The AM Lab, published with their permission. No performance figures are published until their evidence is verified.

The problem

What was actually happening

Manufacturing quotes traditionally require expert manual inspection of incoming CAD files. A senior engineer opens the model, judges whether it can be made, estimates the time and cost, and writes the quote. That expertise is real and scarce, which means the quote queue is exactly as long as one person’s day, and the customer waits for it.

The environment it landed in

  • CAD files in several formats
  • Manufacturing partners with different capabilities
  • Pricing rules held by a few experts
  • Customers comparing response times
  • Accepted orders that must survive into production
An engineer weighing a machined bracket against its CAD model on a laptop, under three cards: CAD review, manual estimate, quote waiting.
Illustrative, not client data
Before

How the work moved before.

Every step below was real work done by a person, and every handoff was a place the process could stop without anyone noticing.

An engineer studying a drawing beside a screen flagging a design revision on the part, with two cards: back to customer, revision requested; quotation on hold.
Illustrative, not client data
What Octacer built

The system that replaced it.

A manufacturing quotation screen with the CAD model beside three checks (manufacturability checked, pricing rules applied, quotation ready) and buttons to accept the quote and move to payment and order.
Illustrative, not client data
  • CAD intake that understands the file

    Uploads are parsed rather than attached, so geometry is available to the rest of the system instead of sitting in an inbox.

  • Manufacturability analysis

    The geometry is checked against what the process can actually produce, and the findings are shown to the customer in the same place they uploaded.

  • Time and cost as rules, not memory

    Pricing logic lives in the platform as data an expert maintains, which is what lets a quote be produced without an expert in the loop for every part.

  • A generated quotation and an order path

    The quote, the acceptance, the payment and the resulting order are one continuous state machine rather than five emails.

  • Three role-scoped surfaces

    Customer, manufacturing partner and internal office each see the same order through the lens of what they are responsible for.

Architecture

The decisions that make it hold.

Not a component list. These are the choices that determine whether the system is still trustworthy a year after launch.

One order, three views: a phone for the customer, a monitor for the internal office and a tablet for the manufacturing partner, each showing the same part and order as paid.
Illustrative, not client data
  • Role-scoped surfaces on one order

    The customer, the manufacturing partner and the internal office are three views of a single record, not three systems kept in step by hand.

  • Analysis as a service

    Manufacturability checks run as their own step, so the rules can change without touching intake or pricing.

  • Pricing rules as data

    Rates, processes and constraints are maintained by the people who own them, and every quote records which version priced it.

  • An order state machine

    Quoted, accepted, paid, in production, fulfilled: named states with defined transitions, so nothing lives between two emails.

  • A partner handoff record

    What was sent to which manufacturing partner, and what came back, is part of the order rather than a separate thread.

Business effect

What changes once it exists

  • A customer finds out whether their part can be made in the same session they uploaded it.
  • Expert judgement is spent on the rules and the hard parts rather than on every routine quote.
  • An accepted quote becomes an order without anyone retyping it.
  • The manufacturing partner works from the order, not from a forwarded email.

Why it matters: relevant for

  • Manufacturing
  • RFQ automation
  • Supplier portals
  • Industrial commerce
  • Quotation systems
Two colleagues at a laptop in front of the shop floor beside a production order panel: quote accepted, payment confirmed, partner assigned, ready for production.
Illustrative, not client data
Next step

Have a similar operational constraint?

Bring the process that looks closest to this one. We map it end to end before proposing anything.