Biz Plan
⏎ Back to Capital Plan
Product Technical Feasibility Reports
Independent-style feasibility validation for predictive systems, ingestion pipelines, curb-scan hardware, and hybrid infrastructure.
PropChain commissioned internal and advisor-supported feasibility assessments—referred to collectively as PropPredict Technical Feasibility and ArchiSpect Technical Feasibility—to ensure that all critical systems in the Seed scope are technically achievable within the Seed capital constraints, on the timeline defined in the Roadmap, and without violating MLS licensing boundaries.
This section synthesizes those analyses into a unified, diligence-ready technical feasibility narrative.
Objective of Feasibility Review
The feasibility work was designed to validate architectural soundness, compatibility with PropChain's hard constraints from the Seed Capital Plan, operational feasibility, modeling feasibility, and Seed-scope alignment.
Architectural Soundness
  • Terra Net curb-scan ingestion
  • Terra Engine predictive intelligence
  • NAS-first + Azure hybrid infrastructure
  • Mobile + backend interaction model
Budget Constraints
  • Cloud budget: $40k/24m
  • Data/MLS budget: $40k/24m
  • Tools: $10k
  • Peak monthly burn: $58–61k base
Operational Feasibility
  • 1,050 curb-miles per rig per month
  • GPS, parcel, MLS alignment
  • Predictive inference under 5 seconds
  • IDX/VOW compliance maintained
All feasibility work concludes that PropChain's path is realistic, cost-contained, and technically executable at Seed.
Technical Feasibility Summary
PropChain's architecture, models, ingestion systems, and mobile experience are technically feasible under the Seed budget, the planned 24-month timeline, the 1→3 rig scanning scenarios, Azure-when-needed scaling, and MLS compliance constraints.
NAS-first architecture keeps cloud costs well within the $40k/24m envelope
Terra Engine predictive stack feasible with existing codebase plus advisor support
CurbScan rig (Jetson AGX Orin) validated for required throughput
GPS→parcel alignment models validated for ORI-2 environments
Hybrid MLS + curb-signal feature store achieves competitive performance
Inference under 5 seconds on target devices is consistently achievable
Seed-level scanning volumes offer enough curb/parcel density for reliable MVP predictions
There were no blockers identified that would require post-Seed capital.
Feasibility — Terra Engine (Predictive Layer)
Modeling Approach Validated
The feasibility reports confirm that PropChain's predictive system—valuation, trends, renovation ROI, and curb/parcel signals—can be built using LightGBM for tabular + MLS-driven features, CNN embeddings for curb-level imagery, hybrid ensembles enabling per-market calibration, and drift-detection layer to maintain accuracy across ORI-2 states.
This aligns fully with the current architecture in Terra Engine.
Data Availability Feasibility
Under Appendix A, MLS access requires S2 Moderate Grow (~$20k/year), CurbScan data from 1–3 rigs generating ~1k curb miles/month/rig, and user/behavioral telemetry from low-volume mobile metadata.
The predictive team concluded that no additional high-cost data sources are required to ship the Seed roadmap.
<5s
Prediction Response
Mobile experience target met consistently
2.4-3.7s
Actual Inference Time
Tests on mid-tier phones with production payloads
25-35%
Cloud Budget Share
Terra Engine infrastructure consumption in base scenario
Training uses Azure spot compute + NAS GPU workload with no need for full-time cloud clusters or GPUs at Seed scale. Block-level valuation error remains within competitive band for early markets, market trend-phase classification is feasible with MLS time-series alone, and curb-score features are demonstrably additive to valuation/ROI models.
Feasibility — Terra Net (CurbScan Ingestion)
01
Rig Hardware Feasibility
The BOM validated that a single Jetson AGX Orin–based rig meets throughput requirements (frame slicing, GPS tagging, compression), consumes ~$13–14k CapEx (consistent with Appendix A), and integrates cleanly with vehicle power and GPU-accelerated workloads.
02
Data Processing Pipeline Feasibility
The ingestion stack—including frame extraction, GPS alignment, curb-signal detection—is computationally feasible on Jetson at edge (light inference), NAS cores (batch ETL), and Azure GPU (training-grade workloads only as needed).
03
Scanned Mile Feasibility
Feasibility confirms the rig can sustain ~50 curb miles/day and ~1,050 miles/month with less than 3% data loss under typical conditions. This directly aligns with all scanning economics in Appendix A.
04
Compliance Feasibility
Critically validated: Curb-scan data is proprietary imagery that does not violate IDX/VOW because it is not MLS-originated, not displayed as listing images, processed for analytics only, and legally separable from MLS data.

Multi-Market Expansion Feasibility: Phases A/B/C scanning (NJ/NY → FL/TX → CA/WA/GA) are feasible without rig redesign. GPS/parcel alignment tests performed on NJ, NY, FL, and CA sample data confirm the pipeline scales across ORI-2 geographies.
Feasibility — Infrastructure (Hybrid NAS + Azure)
Key Feasibility Findings
NAS-first approach avoids high cloud egress and storage costs. Azure required primarily for heavy training jobs, periodic re-calibration, and inference autoscaling during peak demand.
Asset storage, MLS-normalized data, and curb-scan outputs easily fit within the $40k cloud budget and $40k MLS/data budget.
Terraform + Helm templates validated for stable AKS deployment. No vendor lock-in issues identified. Latency acceptable for national GTM.
Compliance Feasibility
  • Hybrid infra allows physical/logical separation of MLS-restricted fields
  • Meets RESO 1.6+ norms
  • Enables MLS-specific data retention schedules
NAS-First Storage
Avoids cloud egress costs while maintaining fast local access
Azure When Needed
Training jobs and peak demand autoscaling only
MLS Separation
Physical and logical isolation of restricted fields
Feasibility — Mobile & Dashboard Experience
React Native Selected
Feasible across iOS + Android with shared codebase and native performance
Predictive Cards Rendering
Under 16 ms/frame for smooth, responsive user experience
Offline Cache Feasible
Less than 40 MB per market for offline functionality
UX Flows Validated
Predictive Search, Map/List Toggle, Curb Insights all tested
Interaction with Predictive Services
Terra Engine Payload
Schemas stable and optimized for mobile delivery
Curb Insights Integration
Terra Net lightweight signals seamlessly integrated
Error States
Degraded mode flows validated for reliability
Dashboard (Pro + Partner) feasible using shared component library, ensuring consistent experience across all user types and platforms.
Risk Assessment & Mitigation
MLS Licensing Delays
Risk: Data availability in initial markets.
Mitigation: Aggregator coverage (ListHub), multi-MLS pipeline, parallel onboarding strategy.
Scanning Program Variance
Risk: Underperformance of rigs in new geographies.
Mitigation: Buffer in scanning OpEx, fallback to partial imagery + extrapolation, redundancy via synthetic curb features.
Model Drift in Volatile Markets
Risk: Accuracy degradation during market volatility.
Mitigation: Drift-detection module, dynamic ensemble weighting, state-level calibration.
Data Storage Pressure
Risk: Storage costs exceeding budget projections.
Mitigation: Tiered NAS retention, compressed curb imagery, selective upload to cloud.
Overall Feasibility Conclusion
All core systems required for Seed MVP are validated as technically achievable
Infrastructure costs remain within Seed budgets
Scanning program feasible at 1–3 rigs
Predictive stack feasible with current data access and modeling approach
No critical blockers identified
Seed runway, burn, and capital constraints remain intact
PropChain's Seed-scope product is technically feasible, operationally executable, and designed to scale directly into Series A with minimal re-architecture.
The comprehensive feasibility validation across all technical domains—predictive modeling, curb-scan ingestion, hybrid infrastructure, and mobile experience—confirms that PropChain can deliver its AI-first, mobile-centric real estate platform within the constraints of Seed capital, timeline, and operational capacity.
Next Steps: Telemetry & Data Capture
With technical feasibility validated across all core systems, PropChain now moves to implement a comprehensive telemetry and data capture plan. This plan ensures that every user interaction, predictive query, and curb-scan ingestion generates actionable intelligence to refine models, optimize UX, and demonstrate product-market fit to Series A investors.
User Behavior Telemetry
Track search patterns, predictive tool usage, and conversion funnels
Model Performance Metrics
Monitor inference latency, accuracy drift, and feature importance
CurbScan Quality Assurance
Validate rig throughput, GPS alignment, and image quality
The telemetry framework will be built into the mobile apps, dashboards, and backend services from day one, ensuring PropChain captures the data needed to iterate rapidly, validate assumptions, and build a compelling growth narrative for the next funding round.

Investor Note: The fully expanded Section 5.5 — Telemetry & Data Capture Plan is available upon request and details the specific metrics, instrumentation architecture, privacy compliance, and KPI dashboards that will drive PropChain's data-informed product evolution.