back_office_ops · realestate · workflow
ZURU improved floor plan generation accuracy by 109% using Amazon Bedrock and Amazon SageMaker
ZURU's Dreamcatcher platform needed to generate 2D floor plans from natural language prompts with high accuracy across two criteria: instruction adherence (understanding room types, purposes, and orientation) and mathematical correctness (dimensions, positioning, and orientation), but the baseline GPT2 model did not meet the required accuracy threshold.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · User enters natural language prompt
A user query such as 'Build me a house with three bedrooms and two bathrooms' initiates the workflow.
Tools used
Amazon BedrockClaude 3.5 SonnetMistral 7BAmazon Bedrock Knowledge BasesAmazon OpenSearch ServerlessAmazon DynamoDBAmazon S3Llama 3.1 8BGPT2LoRA
Outcome
Prompt engineering with Claude 3.5 Sonnet and full fine-tuning with Llama 3.1 8B both achieved 109% improvement in instruction adherence over the baseline GPT2 model; full fine-tuning also delivered a 54% increase in mathematical correctness over the baseline.
Results
Time saved25 hours
Volume109%
Grounding & classification
Source type: technical build writeup
34 fields verified against source quotes, 1 dropped as unverifiable.
content generationragknowledge basefailure mode describedhuman review describedmetric backednamed customertools describedvendor confirmedworkflow describedmanufacturingreal estateaccuracy improvementtechnical build writeupback office opsagentic task executionrag answering