Where we solve AEC’s hardest AI problems in public.
Evidence over promises. Through AECFoundry Labs we share our research, test our ideas, commercialize products, and gather industry feedback that informs platform development. A way to collaborate with the AEC industry at large and contribute back to advancing AI.
Research
We’d rather show the rigor than claim a score.
We topped the first independent benchmark for AEC document tasks, and we’ll tell you why a leaderboard is still the wrong test. Public benchmarks reward verbose answers over correct ones. So we built a stricter one of our own, and the real evaluation we run for every client is a private benchmark on their documents.
Research
AECV-Bench
The first benchmark validating multimodal AI on real architectural drawings. We tested 12 frontier models on object counting, OCR, and spatial measurement — evidence, not promises.
Explore →
Publication
Agentic Safety Systems for Construction
This research investigates how domain-specific multi-agent generative artificial intelligence systems can automate construction safety risk assessments and subsequently share them with knowledge graphs (KGs) to enable downstream reuse of the extracted knowledge and overcome information isolation.
Explore →
Publication
AI Agentic System for Safety Risk Assessments
We introduce a safety risk assessment platform leveraging agentic AI systems, LLMs, hybrid semantic search, and RAG to generate expert-level safety risk assessments. Presented at CONVR 2024 and awarded Best Presentation.
Explore →
Products
Products, tools, and prototypes. On one foundation.
Each one tests a different slice of our capabilities against real-world data, all running on the same foundation. This is how we continuously learn and improve what we deliver at enterprise level.
Product
DocumentFoundry
Parse. Extract. Understand. Turns construction drawings, specs, and project documents into structured, auditable data. Powering QA/QC, compliance review, and downstream automation.
Explore →
Product
Archie
Your AI copilot for building codes and standards. Search, navigate, understand, and apply codes in plain language with cited answers — and analyse documents, models, and drawings for compliance.
Explore →
Product
SWMS AI
Job-specific safety risk assessments in seconds. AI trained on industry safety data drafts the work sequence, identifies hazards, assigns risk scores, and suggests control measures.
Explore →
Built on the same foundation.
Our research and products are built on and inform the same foundational platform and AEC-specific pipelines we deploy and configure for clients. Bring us a workflow and we’ll show you on your own documents.