Construction Tech / AEC SaaS / AI Compliance

GPT-4o + RAG - 200+ code checks across five disciplines, every finding cited to a real FBC/IBC/NFPA section. Days of manual review cut to supervised hours.

AI Building-Code Compliance & Permit-Review SaaS

An AI-powered construction compliance platform that automates drawing reviews, verifies building codes, and generates audit-ready engineering reports with accurate citations.

Overview

Permit review means reading hundreds of pages of construction drawings against hundreds of code rules by hand. For AEC firms and Florida code-engineering practices, that process takes days per project, requires deep discipline expertise across Architectural, Structural, Mechanical, Electrical, and Plumbing trades, and leaves no structured audit trail for the permit file.

This platform cuts that from days to hours - and, critically, makes the AI auditable. Every finding the system raises is grounded in a real code passage retrieved by ChromaDB vector search from indexed FBC, IBC, and NFPA standards. The AI cannot invent a rule: it cites, it does not guess. Reviewers verify reasoning instead of trusting a black box, and the output is a fully cited, permit-ready compliance report in both PDF and JSON.

Built on FastAPI + Celery + PostgreSQL backend, Next.js 15 frontend, and a six-stage transparent processing pipeline with live progress, per-stage cost display, and a hard billing ceiling per document. Live in production for a Florida building-code engineering firm.

The platform ingests construction PDF drawings, runs AWS Textract OCR and GPT-4o per-page extraction, retrieves the exact relevant code passage via ChromaDB vector search (RAG), and runs 200+ checks per discipline across five trades: Architectural, Structural, Mechanical, Electrical, and Plumbing, plus NFPA fire and life-safety codes (NFPA 1, 70, 72, 101).

Every finding links to a real citation - for example, FBC Plumbing 2023, Section 712.1 - so the AI cannot hallucinate a rule. It must reference an indexed passage from the FBC, IBC, or NFPA standards stored in ChromaDB. The output is a fully cited, audit-ready compliance report in PDF for the permit file and JSON for BIM/PM tooling, with every page traceable to the source drawing.

The system is cost-aware by design: page, OCR, and LLM cost is estimated up front with a hard ceiling per document to prevent runaway spend. Reviewers operate in a full human-in-the-loop workflow - they can edit any extracted field and re-run checks before accepting the final report.

- Relevant keywords -

Challenge & Solution

Solving real problems with smart engineering

Six-Stage Transparent Processing Pipeline

Problem

Manual permit review takes days per project - reading hundreds of pages of construction drawings against hundreds of code rules by hand is the standard process at AEC firms and code-engineering practices. A single project review can take two to five days of senior engineer time, with no structured tracking of which rules were checked or why a finding was raised.

Solution

Documents move through a fully visible pipeline: Queued → OCR (AWS Textract) → Schedule Detection → Table Extraction → Equipment Parsing → Ready-for-Verification. Each stage shows live progress and its own cost, so reviewers always know exactly where processing is and what it is costing.

Audit-Ready PDF + JSON Export

Problem

No audit trail for the permit file - traditional review produces a marked-up PDF or a handwritten checklist, neither of which gives the permit office a structured, traceable record of which code sections were checked, which passed, and which failed. When a drawing is revised, the review process starts from scratch with no reference to what was previously verified.

Solution

Every report exports as a printable PDF for the permit file and a complete JSON record for BIM and PM tooling. Every finding links to its source drawing page, the extracted field, and the specific code citation - creating a full traceable chain from drawing to code section.

Zero-Hallucination RAG-Grounded AI

Problem

AI hallucination makes LLMs unusable for compliance without RAG - off-the-shelf LLMs cannot be trusted for code compliance because they generate plausible-sounding but unverified code citations. A building official cannot accept 'AI says this complies' without a reference to the actual code section. Without grounding every finding in a retrieved passage from an indexed code corpus, AI is a liability rather than an asset in permit review.

Solution

Every finding is grounded in a real code passage retrieved by ChromaDB vector search from indexed FBC, IBC, and NFPA standards. The AI cannot invent a rule; it must cite an indexed passage before raising any flag. This makes the output permit-ready and professionally defensible, not just a plausible LLM response.

200+ Discipline-Specific Code Checks Across Five Trades

Problem

Five disciplines require different expertise and rule sets - Architectural, Structural, Mechanical, Electrical, and Plumbing reviews each require a different body of knowledge, different sections of the FBC, IBC, and NFPA standards, and different extraction logic from construction drawings. A single generic AI cannot handle all five trades without discipline-specific prompts, rule sets, and structured output schemas.

Solution

Each discipline (Architectural, Structural, Mechanical, Electrical, Plumbing) gets its own prompts, extraction logic, rule set, and check library. Plumbing checks drainage slope, backflow prevention, and venting. Structural checks load paths, framing, and foundations. Electrical checks NEC compliance. Every check is traceable to a specific code section.

Streaming 500 MB Uploads with Deduplication

Problem

Large PDF processing is expensive and unpredictable - construction drawing sets can be hundreds of pages and hundreds of megabytes. Sending entire documents to an LLM is cost-prohibitive, technically unreliable, and produces unstructured outputs that cannot be audited. Without per-page processing, cost control, and structured JSON extraction, AI-assisted review is not viable at production scale.

Solution

Construction drawing PDFs are streamed to disk in 5 MB chunks with magic-byte validation and SHA-256 content hashing. Duplicate submissions are detected and rejected before any processing begins - preventing double-billing on resubmitted drawings and ensuring storage integrity.

Per-Page Structured GPT-4o Extraction with Confidence Scores

Problem

No human-in-the-loop means no professional accountability - code-compliance decisions carry professional and legal liability. A system that produces a report without allowing a licensed engineer to review, edit, and approve every extracted field and finding is not usable in a regulated permit-review workflow. Reviewers need to verify AI reasoning, not accept its output as final.

Solution

GPT-4o processes each page individually and outputs typed JSON with confidence scores for every extracted field. Reviewers can inspect, edit, and approve extraction results before checks run - full human-in-the-loop at the data layer, not just the output layer.

Cost-Aware Processing with Hard Ceilings

Problem

Billing surprises kill production AI tool adoption - AEC firms and code-engineering practices cannot adopt an AI tool that runs unpredictable LLM costs per document. Without up-front cost estimates, live per-stage cost display, and a hard ceiling per document, the tool is a financial risk rather than a productivity gain.

Solution

Page count, OCR cost, and LLM token cost are estimated up front before any processing begins. A hard cost ceiling per document prevents runaway spend. Live cost display per stage and per page gives practices complete financial visibility into every review job.

Features

What it does

Document Ingestion & Deduplication

Streaming upload handler for PDFs up to 500 MB - 5 MB chunk streaming to disk
Magic-byte MIME type validation - rejects non-PDF files regardless of extension
SHA-256 content hash deduplication - duplicate submissions rejected before any processing or billing
S3 storage with signed URL access control - no public document access
Cost estimate displayed and acknowledged before any Celery job is queued

OCR Pipeline (AWS Textract)

Per-page Textract processing returning structured blocks: lines, words, tables, key-value pairs
Block output stored as JSONB in PostgreSQL per page for downstream extraction stages
Textract cost tracked per page and accumulated against the per-document hard ceiling
Schedule and title block detection identifies drawing metadata for cross-referencing
Table rows parsed into typed JSON for equipment schedules and specification tables

AI Extraction (GPT-4o + PydanticAI)

Per-page structured extraction: typed JSON output with confidence scores for every field
Discipline-specific prompts for Architectural, Structural, Mechanical, Electrical, and Plumbing pages
Equipment parsing: mechanical schedules, panel schedules, fixture schedules, structural member tables
PydanticAI validates every GPT-4o output before database write - no unstructured LLM responses stored
Low-confidence fields automatically flagged for mandatory reviewer correction

RAG-Grounded Compliance Checks (ChromaDB)

FBC, IBC, and NFPA 1/70/72/101 standards indexed as OpenAI vector embeddings in ChromaDB
Top-k code passage retrieval before every GPT-4o compliance check - grounds every finding
200+ discipline-specific checks: Architectural (occupancy, egress, accessibility), Structural (loads, framing, foundations), Mechanical (HVAC, ventilation), Electrical (NEC panels, circuits), Plumbing (drainage, backflow, venting)
NFPA fire and life-safety checks: NFPA 1 (Fire Code), NFPA 70 (NEC), NFPA 72 (Fire Alarm), NFPA 101 (Life Safety Code)
Every finding includes: code citation, retrieved passage excerpt, extracted drawing value, severity, and source page

Human-in-the-Loop Review

Per-page extraction review UI: inspect and edit any GPT-4o extracted field before checks run
Per-finding approval: accept, reject, or flag each compliance finding with reviewer comments
Reviewer corrections stored separately from raw AI output - checks always run against approved data
Full audit log: every edit, approval, and rejection timestamped with reviewer identity
Re-run checks on any corrected extraction without reprocessing the full pipeline

Cost Control & Observability

Up-front cost estimate (page count + Textract + LLM tokens) before processing begins
Live per-stage cost display on the pipeline progress UI
Configurable hard cost ceiling per document - enforced by Celery task gating
Cost ceiling alert via Sentry notification if threshold is reached mid-pipeline
Sentry error tracking + structlog JSON logs with per-request trace IDs across all pipeline stages

Report Generation (PDF + JSON)

Branded PDF compliance report rendered by ReportLab for the permit file
Every finding links to: source drawing page, extracted field value, retrieved code citation, reviewer approval
Complete JSON export for BIM and PM tooling with full finding metadata
Both outputs stored to S3 with signed URL download links
Immutable report versions: any post-approval change creates a new versioned report record

Processing Pipeline UI

Six-stage progress view: Queued / OCR / Schedule Detection / Table Extraction / Equipment Parsing / Ready
Per-stage status indicators with timing and cumulative cost display
TanStack Query polling for live async Celery pipeline updates without page refresh
Document history view: all past submissions with processing status, cost totals, and report access
Duplicate submission alert: shows original submission date and cost if SHA-256 match detected

Tech Stack

Built with

Web Console (Next.js)

Next.js 15 + React 19 - reviewer console
TypeScript - end-to-end type safety
Tailwind CSS - styling
TanStack Query - live pipeline polling
Sentry - client-side error tracking

Backend (FastAPI)

FastAPI on Python 3.11 - async API with uvicorn + gunicorn
Pydantic v2 - typed schemas and validation
PydanticAI - structured LLM outputs
async SQLAlchemy + asyncpg + Alembic - ORM and migrations

Processing Pipeline

Celery - six-stage document pipeline as task chains
Redis - broker and result backend
Horizontal Celery scaling - throughput on demand
Per-stage progress and cost tracking

OCR & Extraction

AWS Textract - per-page OCR blocks, tables and key-value pairs
Per-page JSONB storage in PostgreSQL
Structured extraction with confidence scores

AI Engine

OpenAI GPT-4o - vision and structured output per page
PydanticAI - model routing and output validation
OpenAI text-embedding-3-small - embeddings for retrieval

RAG & Vector Store

ChromaDB - FBC, IBC and NFPA 1/70/72/101 embeddings
OpenAI Embeddings - code-passage indexing
Top-k retrieval per compliance check - every finding cited

Database

PostgreSQL - primary store
JSONB columns - OCR blocks, extractions and findings
asyncpg + Alembic - async driver and migrations

Storage & Security

AWS S3 - PDFs, OCR outputs and reports
Pre-signed multipart upload - 500 MB streaming
Signed-URL access control with configurable expiry
Per-reviewer audit logging

Observability & Reporting

Sentry - error tracking across backend and frontend
structlog - JSON logs with trace IDs
Per-stage cost logging to PostgreSQL
ReportLab - branded compliance report PDFs

DevOps & Infrastructure

Docker - containerised builds
GitHub Actions - CI/CD
Coolify on DigitalOcean - managed Docker Compose
Traefik - reverse proxy
Cloudflare + Let’s Encrypt - DNS and TLS

Results & Value

What actually changed

Delivered a production AI permit-review platform (Next.js 15 SPA, FastAPI backend, six-stage Celery pipeline) for a live Florida code-engineering firm
Zero-hallucination compliance findings: ChromaDB RAG retrieves real FBC/IBC/NFPA citations before every GPT-4o check - the AI cites, it does not guess
200+ discipline-specific checks across Architectural, Structural, MEP (Mechanical, Electrical, Plumbing), and NFPA fire/life-safety codes in one codebase
Full human-in-the-loop workflow: per-page extraction review, per-finding approval, and immutable report versioning for professional accountability
Cost-aware architecture: up-front estimates, live per-stage tracking, and configurable hard ceilings prevent billing surprises at production document volumes
Streaming 500 MB upload with SHA-256 dedup and signed S3 URLs - production-grade data handling for large construction drawing sets
End-to-end observability: Sentry + structlog across backend and frontend with per-request trace IDs and per-stage cost logging to PostgreSQL
Horizontal Celery scaling with zero code changes - worker replicas add capacity without architectural changes