Skip to content
AI ConnectPowered by VELENTIS
AI-assisted1 min

ONESTRUCTION and AWS Develop Construction-Focused Ishigaki-IDS Foundation Model

ONESTRUCTION worked with the AWS Generative AI Innovation Center to build Ishigaki-IDS, a specialized foundation model designed for construction and BIM workflows.

(KI-generiertes Symbolbild: Gemini / AI Connect)

ONESTRUCTION built a specialized technical system known as Ishigaki-IDS with technical advisory from the AWS Generative AI Innovation Center. The resulting development produced a foundation model specialized for construction and BIM workflows. This architectural case study demonstrates how technical teams can build a domain model in a data-scarce field.

The underlying training methodology relied on specific architectural components deployed on cloud infrastructure. According to the case study, the team combined synthetic data, a three-stage training pipeline, and verifiable rewards on Amazon EC2. These techniques allowed ONESTRUCTION to address domain requirements within specialized industrial workflows.

What this means for you

The architectural approach highlights how organizations can train customized foundation models even when domain-specific training data is limited. Combining synthetic data with multi-stage pipelines and verifiable rewards on Amazon EC2 offers a blueprint for building targeted AI tools for specialized fields like BIM workflows.

Evidence

Solidly sourced
46/100

  • ONESTRUCTION built Ishigaki-IDS with technical support from AWS as a model tailored for construction and BIM tasks.

    single source
    Quote

    „built Ishigaki-IDS, a foundation model specialized for construction and BIM workflows.“

    AWS, Machine Learning

  • The development pipeline utilized synthetic data, a three-stage process, and verifiable rewards on Amazon EC2 to overcome data scarcity.

    single source
    Quote

    „combined synthetic data, a three-stage training pipeline, and verifiable rewards on Amazon EC2 to build a domain model in a data-scarce field.“

    AWS, Machine Learning

Sources

The evidence score is computed, not hand-set: from confidence, the number of sources and the share of verified statements.

Source & transparency

As of: August 11, 2026

AI-assistedAI-assisted, editorially reviewed

Sources
1
Verified statements
0 / 2
Evidence score
46Solidly sourced
Generative AIUse Cases

Want to put this into practice?

We connect you with suitable AI providers from the DACH region, free of charge and without obligation.

What's next?

What does AI really cost?

Honest price ranges for your project, independent, with nothing to sell you.

See prices

Get your team ready (AI Act)

Since February 2025 the EU AI Act requires measures to promote AI literacy. Baseline training for your team is in development.

Find the right training

Find the right provider

Describe your project. We connect you with suitable providers at a fair price.

Start the analysis
Back to AI Radar