Frequently Asked Questions
Honest answers about methodology, validation status, limitations, and what makes PROTHEUS different.
💡 Our Promise: Radical Transparency
We don't hide limitations or inflate claims. If we don't know something, we say so. If we're working on it, we share the roadmap. Trust is earned through honesty, not marketing hype.
Core Methodology
Understanding how PROTHEUS works
How is PROTHEUS different from ChatGPT/Copilot estimates?
PROTHEUS is 90% deterministic + 10% LLM. Here's the breakdown:
PROTHEUS APPROACH:
✓ Hardcoded validation constraints
✓ Multi-factor viability model
✓ Knowledge base (patterns, rates)
✓ Regional market rates
✓ Location-adjusted buffers
✓ Strategic LLM use (architecture, stories)
→ Explainable + reproducible
ChatGPT/COPILOT APPROACH:
✗ Pure LLM guessing (no rules)
✗ No validation logic
✗ No market rate data
✗ No regional adjustments
✗ No compliance detection
✗ Hallucinations common
→ Black box + inconsistent
Key difference: PROTHEUS shows you exactly why it recommends adjusting timeline or budget. ChatGPT just provides vague estimates with zero justification.
Example:
Input: "Healthcare app, small team, compliance requirements, tight timeline, limited budget"
ChatGPT: "Looks feasible, maybe add some buffer"
PROTHEUS: "NOT FEASIBLE. Significant budget gap detected. Compliance requirements add substantial implementation time. Realistic timeline and budget provided with detailed breakdown. Viability: MODERATE-HIGH RISK."
What are the validation constraints exactly?
Hardcoded, immutable checks that catch fatal flaws before you write code:
Technology Fit:
• Stack appropriateness for project tier
• Architecture-stack compatibility
Resource Alignment:
• Timeline feasibility with learning curves
• Budget-cost balance validation
• Team skills match assessment
• Budget capacity threshold checks
Completeness & Security:
• Feature completeness for domain
• Security baseline enforcement
• Setup complexity validation
Zero LLM hallucination in these constraints. They're encoded from 25+ years of architecture experience. See full methodology for details.
How do you calculate budgets and timelines?
Budget formula (transparent):
Example (E-commerce MVP):
Story points calculated from feature breakdown
× Regional market rate
× Coordination overhead
+ Infrastructure (hosting platform)
+ Compliance requirements
= TOTAL with confidence interval
Timeline formula:
All methodology documented in Methodology page.
Validation & Accuracy
Honest disclosure about what's validated and what's not
Is PROTHEUS validated with real project outcomes?
Honest answer: Not yet. Here's our current validation status:
✅ What IS Validated:
- • Internal test suite: 85-95% pass rate (comprehensive automated testing)
- • Validation constraints: Deterministic checks (100% reproducible)
- • Budget calculations: Market rates verified (industry sources)
- • Knowledge base: Verified patterns and common issues
❌ What Is NOT Validated:
- • Timeline predictions vs actual delivery dates (requires retrospective data)
- • "500+ projects analyzed" (we don't have this corpus)
- • Specific accuracy percentages (cannot claim without real outcome validation)
- • Budget estimates vs actual costs (need client follow-up data)
Why we're honest about this: Many AI tools claim "validated with thousands of projects" without proof. We refuse to make unverified claims. We'll earn validation through real data in 2025.
Validation Roadmap:
Q1 2025 (Now): Launch with honest messaging, collect early adopters
Q2 2025: Follow up with early adopters, compare predictions vs reality
Q3 2025: Comparative analysis with multiple validation approaches
Q4 2025: Publish validated accuracy report with completed projects
How accurate are PROTHEUS predictions?
Honest answer: We don't know yet—we haven't validated against real project outcomes. Here's what we can say:
What we measure internally:
• Test suite pass rate: 85-95% (comprehensive automated validation)
• Constraint validation: 100% (deterministic checks execute consistently)
• Budget calculation consistency: 100% (reproducible formulas)
• Quality validation: Multiple automated and objective checks
What we'll measure in 2025:
• Timeline prediction accuracy vs actual delivery
• Budget estimation accuracy vs actual costs
• Risk detection effectiveness for project blockers
• Viability score correlation with project outcomes
Why this matters: We could claim high accuracy right now (many tools do), but it would be meaningless without real validation. We prefer to earn trust through transparency than inflate claims.
How does the analysis process work step-by-step?
PROTHEUS follows a systematic 4-phase pipeline:
Context Loading
Loads relevant domain knowledge, market rates, and technical constraints from our knowledge base.
Multi-Agent Validation
9 deterministic agents analyze feasibility, budget, timeline, risks, and technical requirements in parallel.
Generative Enhancement
3 LLM agents create architecture designs, user stories, and opportunity analysis using strategic AI calls.
Quality Assurance
13 automated tests validate output quality, then results are stored in the knowledge base for continuous improvement.
Typical analysis time: 2-5 minutes for complete report generation.
How does PROTHEUS compare to traditional project estimation methods?
PROTHEUS combines the best of expert analysis with AI efficiency.
Traditional Methods:
Expert Estimation:
• High accuracy but expensive ($500-2000)
• Takes 1-2 weeks
• Limited scalability
• Subject to expert availability
Analogous Estimation:
• Uses past project data
• Requires historical database
• Limited to similar projects
• Manual process
PROTHEUS Advantages:
Speed & Cost:
• 5 minutes vs 1-2 weeks
• Free tier available
• 24/7 availability
• No expert scheduling needed
Consistency & Transparency:
• Same input = same output
• Detailed reasoning provided
• Validation constraints applied
• Market data integration
Best Use Case:
Use PROTHEUS for initial screening, then consult experts for complex projects. PROTHEUS catches 80% of feasibility issues before expert consultation.
Technical & Integration
API access, data handling, and technical capabilities
Can I integrate PROTHEUS via API?
Yes, REST API available. Currently in private beta (launching publicly Q2 2025).
Authentication:
API key-based authentication
Core Endpoints:
POST /api/v1/analyze - Submit project for analysis
GET /api/v1/analysis/:id - Retrieve results
GET /api/v1/exports/:id/pdf - Download PDF report
Rate Limits:
Tier-based limits aligned with subscription plans
Webhooks:
Callback notifications when analysis completes (Q2 2025)
Interested in API access? Email api@protheusai.com for early access.
Do you store my project data?
Analysis metadata only (anonymized). Source code never stored.
What we store:
- • Project brief (text description)
- • Analysis results (ADRs, user stories, budget, timeline)
- • Anonymized metadata (stack, team size, domain, region)
What we DON'T store:
- • Source code (never uploaded, never stored)
- • Credentials (API keys, passwords, secrets)
- • Client names (unless you explicitly provide them)
- • Proprietary business logic
Data retention:
Configurable retention period, then auto-deleted.
Export options: PDF, Markdown, JSON (download before expiration)
Compliance:
GDPR-compliant. EU data residency available.
Data Processing Agreement (DPA) available on request.
See full Privacy Policy andSecurity page.
What if my project uses a niche stack?
PROTHEUS flags high uncertainty and provides wider confidence intervals.
Knowledge Base Coverage:
• Popular frameworks and platforms
• Known technical conflicts and compatibility issues
• Regional market data across multiple regions
• Compliance requirements (GDPR, HIPAA, PCI-DSS, SOC2, WCAG)
If your stack is truly novel:
PROTHEUS will flag: ⚠ HIGH UNCERTAINTY
Confidence intervals widen appropriately
Recommendation: "Stack validation recommended before committing budget"
How we handle edge cases:
1. Check knowledge base for similar patterns
2. Apply general heuristics for novel technologies
3. Increase confidence intervals appropriately
4. Recommend human architect review
Can PROTHEUS handle microservices architectures?
Yes, but only for appropriate project complexity tiers. MVPs should typically use monoliths.
Architecture Appropriateness Validation:
Microservices overhead detected:
• Inter-service communication complexity
• Integration testing requirements
• Deployment orchestration needs
• Observability infrastructure
• Typical overhead: Substantial increase vs monolith
Example:
Monolithic approach: Standard timeline and budget
Same features with microservices: Significantly increased complexity
→ PROTHEUS asks: "Is the complexity justified?"
Compliance & Limitations
What PROTHEUS can and cannot do
What about regulatory compliance (GDPR, HIPAA, SOC2)?
PROTHEUS detects compliance requirements and estimates implementation costs.
Auto-detected compliance triggers:
Estimated implementation overhead:
⚠ Important Disclaimer:
PROTHEUS estimates technical implementation costs, not legal/audit costs. Always consult compliance experts for healthcare/finance projects.
What can PROTHEUS NOT do?
Radical transparency: Here's what PROTHEUS cannot do (yet):
❌ Replace Architects
PROTHEUS augments architects, doesn't replace them. Human judgment required for final decisions.
❌ Generate Executable Code
PROTHEUS analyzes pre-code phase (architecture, planning). Use Cursor/Copilot for code generation.
❌ Guarantee Project Success
PROTHEUS reduces risk, doesn't eliminate it. Success depends on execution, not just planning.
❌ Analyze Legacy/Maintenance Projects
Optimized for greenfield (new builds). Brownfield analysis coming Q3 2025.
❌ Handle Very Small Budgets
Minimum analysis threshold applies. Below that, too many unknowns for reliable prediction.
❌ Work with Non-Technical Projects
Requires software engineering scope (web, mobile, API, infrastructure only).
See full Scope & Limitations section.
Do you support non-English projects?
Input: English only (currently). Output: Multiple languages available.
Current Support:
• Analysis input: English only (project brief, requirements)
• Output export: English, Spanish, French (PDF/Markdown)
• Regional data: Multiple regions supported
Roadmap:
• Q2 2025: Additional language input support
• Q3 2025: Multi-language user stories generation
Roadmap & Updates
What's coming next
How often is PROTHEUS updated?
Continuous updates to knowledge base. Major releases quarterly.
Knowledge Base Updates (Regular):
• New patterns and anti-patterns identified
• Market rate updates from industry sources
• Dependency compatibility checks
Validation Logic Updates (Periodic):
• Constraint refinements based on feedback
• New domain patterns added
• Regional factor adjustments
Major Releases (Quarterly):
• Q1 2025: Launch with core validation pipeline
• Q2 2025: API public beta, validation testing
• Q3 2025: Enhanced features and analysis
• Q4 2025: Validated accuracy reporting
What's on the roadmap for 2025?
Q1 2025 (Now) - Launch:
- ✓ Core validation pipeline
- ✓ Free tier availability
- ✓ Honest, transparent messaging
- ✓ Full methodology documentation
Q2 2025 - Validation & API:
- • Public API beta launch
- • Comparative validation testing
- • Additional language support
- • Webhook integration
- • Early adopter outcome tracking
Q3 2025 - Enhanced Analysis:
- • Legacy project analysis capability
- • Enhanced compliance detection
- • Team collaboration features
- • Custom validation options
Q4 2025 - Validated Accuracy:
- • Publish validated accuracy data
- • Enhanced risk assessment
- • Additional platform integrations
- • Mobile application
Still have questions?
💡 Pro Tip: Join our Early Adopter Program to help validate PROTHEUS accuracy and get extended free access.Learn more →
Ready to Try PROTHEUS?
3 free analyses per month. No credit card required. See exactly how PROTHEUS thinks.
Start Your Free Analysis →Full transparency • Multi-agent pipeline • Architectural validation • Strategic LLM use