DiManEx accelerates additive suitability reviews by turning drawings into structured insights with Werk24

Thousandsof drawings triaged in hours

5–20stypical processing time per page

JSONdeterministic schema + confidence scores

Portrait of Henk Jonker, Co-Founder of DiManEx

“With Werk24, processing thousands of technical drawings can now be done in just a few hours.”

— Henk Jonker, Co-Founder, DiManEx

The challenge

DiManEx helps enterprises keep machines running by producing spare parts on demand. To decide what can be additively manufactured — and at what risk — engineers first need to read every drawing: dimensions, GD&T, tolerances, surface finishes, materials, and notes.

Volume & Variability: Thousands of PDFs and scans, each with different conventions
Manual Bottleneck: Triage by hand slowed down AM feasibility reviews
Hidden Risk: Missed tolerance or finish details could derail downstream AM planning

The team needed a reliable way to transform drawings into structured, machine-readable data — fast, and at scale.

Why now

AM programs Scaling: More parts to screen, more stakeholders to inform
Service Levels: Customers expect rapid, data-backed suitability decisions
No Templates to Maintain: Solution must work on mixed-quality scans, not just perfect PDFs

The Werk24 solution

Purpose-built extraction for mechanical drawings

Werk24 reads 2D component & assembly drawings — including acceptable scans and clean hand-drawn engineering documents — and returns structured JSON with field-level confidence scores. It extracts title-block metadata and feature data (dimensions, upper/lower tolerances, GD&T, threads, surface roughness) and normalizes units/symbols.

Deterministic output or a custom format matching DiManEx’s pipeline fields
Insights for AM: external dims, volume estimate, primary/secondary processes (DIN 8580), feasibility flags
No training data required to get started; optional custom quality gate if you want 99% on specific features

Integration that fits the stack

DiManEx consumes Werk24 via HTTPS (webhooks) and Python. Confidence thresholds automatically accept routine fields and route edge cases for review. For faster rollout, Werk24 delivered a custom schema mapping so payloads fill the existing AM analysis models without middleware.

REST API + webhooks for lights-out processing
Python SDK for pipelines and batch runs
Data residency: EU or US; higher tiers offer a zero-retention option

Implementation in Days, not Months

Day 1–2: trial key, sample drawings, output schema agreed
Week 1: first workflow live (AM triage), confidence thresholds tuned
Week 2+: expansion to volume estimates & process hinting, optional quality gate for critical features

No labeled training data required — unless you request a custom quality gate.

Results

Hours → Minutes

AM triage at scale

Thousands of drawings are processed in just hours.

5–20s

Typical per-page latency

Parallel jobs available on provisioned & enterprise tiers.

↓ rework

fewer late surprises

Feasibility flags highlight risky tolerances/finishes early.

EU / US

data residency

Zero-retention option on higher tiers; encrypted in transit & at rest.

From developer: inside for developers: inside

Clean APIs, predictable JSON, confidences and webhooks. From PDF to production in hours instead of weeks.

            pip install werk24    # Install Werk24
werk24 init           # Obtain a trial license
werk24 health-check   # Verify connection

            from werk24 import read_drawing_sync, get_test_drawing, AskMetaData
print(read_drawing_sync(get_test_drawing(), [AskMetaData()]))

Tip: Use webhooks/callbacks for dark processing and confidence thresholds.

            {
    "ask_version": "v2",
    "ask_type": "META_DATA",
    "page_type": "COMPONENT_DRAWING",
    "identifiers": [
        {
            "reference_id": 0,
            "language": null,
            "value": "WK2401",
            "identifier_type": "DRAWING_NUMBER",
            "stakeholder": null,
            "period": null
        }
    ],
    "designation": [
        {
            "reference_id": 1,
            "language": null,
            "value": "ADAPTER"
        }
    ],
    "languages": [
        "ENG"
    ],
    "general_tolerances": null,
    "general_roughness": {
        "reference_id": 2,
        "label": "\u221a",
        "confidence": null,
        "standard": "ISO 1302:2002",
        "machining_allowance": null,
        "material_removal_type": "UNSPECIFIED",
        "applies_all_around": false,
        "direction_of_lay": null,
        "manufacturing_process": "",
        "conditions": [],
        "waviness": null
    },
    "material_options": [
        {
            "reference_id": 3,
            "material_combination": [
                {
                    "raw_ocr": "20MnCr5",
                    "standard": "DIN EN 10084",
                    "designation": "20MnCr5",
                    "material_category": [
                        "FERROUS_ALLOY",
                        "STEEL",
                        "STRUCTURAL_OR_CONSTRUCTIONAL_STEEL"
                    ]
                }
            ]
        }
    ],
    "weight": {
        "reference_id": 4,
        "value": "1.025",
        "unit": "kilogram"
    },
    "projection_method": {
        "reference_id": 5,
        "projection_method": "FIRST_ANGLE"
    },
    "bill_of_material": null,
    "unit_systems": [
        {
            "reference_id": 6,
            "unit_system_type": "METRIC"
        }
    ]
}

            {
    "ask_version": "v2",
    "ask_type": "FEATURES",
    "page_type": "COMPONENT_DRAWING",
    "dimensions": [
        {
            "reference_id": 0,
            "label": "12 ±0.1",
            "confidence": {
                "score": "0.96"
            },
            "quantity": 1,
            "size": {
                "value": "12",
                "unit": "mm",
                "size_type": "LINEAR",
                "tolerance": {
                    "tolerance_grade": "12",
                    "deviation_lower": "-0.1",
                    "deviation_upper": "0.1",
                    "fit": null,
                    "is_theoretically_exact": false,
                    "is_reference": false,
                    "is_general_tolerance": false,
                    "is_statistical": false
                }
            }
        }
    ]
}

Extract shortened - use the SDK for complete processing.

Read API documentation Request the API test key Test online

Explore more case studies

Ready to see it on your drawings?

Try the online demo or request a trial API key (100 drawings).

Try it online Get a trial API key