AI case study

State Inspectorate of Architecture and Urban Development (DIAM)Document verification

Inspectors manually reviewed thousand-page construction applications. Now, AI flags code inconsistencies for final human review.

Published

Key results

Target Processing Time
2 hours
vs 10 working days

Result highlights

Unlock 1 result highlight

The story

Context

A government urban development inspectorate processes construction applications containing hundreds or thousands of pages of drawings, explanatory notes, and certificates.

Challenge

Inspectors must manually review these massive document packages under tight deadlines. Even minor inconsistencies, such as a missing file or a...

Solution
Unlock full story

Quotes

Sources found with NewsCatcher's Web Search API

The company

State Inspectorate of Architecture and Urban Development (DIAM) logo

State Inspectorate of Architecture and Urban Development (DIAM)

diam.gov.ua

Central executive body for architectural and construction control in Ukraine.

IndustryGovernment & Public Sector
LocationKyiv, Ukraine
Employees251-1K
Founded2020

The implementation partner

Role in this case study

Developed the 'Digital Inspector' AI system for DIAM to verify construction project documentation.

IndustryTechnology
LocationOslo, Norway
Employees251-1K
Founded1999

The vendor

De Novo logo

De Novo

de-novo.ua
ROLE IN THIS CASE STUDY

Provided cloud infrastructure with NVIDIA GPUs and architectural expertise for the 'Digital Inspector' project.

IndustryTechnology
LocationKyiv, Ukraine
Employees51-250
Founded2008

Use case

State Inspectorate of Architecture and Urban Development (DIAM)'s Document verification is part of this use case:

Compliance Review
71 case studies(+24% YoY)
Proven impact?
Average
2.8Typical
2.8Typicalwithin Government & Public Sector
2.8Typicalwithin Operations

Similar Case Studies

Related implementations across industries and use cases

Republic of Tajikistan logoR

Republic of Tajikistan

Government & Public Sector|Enterprise

Licensing automation

AgenticL3

Paper applications took 14–30 days; agents now extract, validate, and route each file to a decision in 90 seconds.

50-60%Repetitive Workload Cut
via nvidia.com
Published Oct 23, 2026
European Patent Office logoE

European Patent Office

Government & Public Sector|Enterprise

Document transcription

AgenticL1

Complex math and chemical structures bottlenecked transcriptions. A fine-tuned OCR model now structures this data into ST36 XML.

Few minProcessing Lead Timevs up to 5 days
via mistral.ai
Published Mar 11, 2026
City of Antibes logoC

City of Antibes

Government & Public Sector|Mid-size

Budget compliance

AgenticL2

General AI failed to parse 6,000 complex budget codes. A custom model now maps every line to green goals, cutting manual work by 30%.

30%Manual Workload Reduced
via sap.com
Published Dec 1, 2025
Judicial Power of the State of Querétaro logoJ

Judicial Power of the State of Querétaro

Government & Public Sector|Mid-size

via noticiasdequeretaro.com.mx
Unlock to view details

71 AI case studies in Compliance Review

121 AI case studies in Government & Public Sector

362 AI case studies in Operations