DealSphere Phase 1 (MVP) — Product Requirements Document¶
Table of Contents¶
- Product Overview
- Target Users
- Core Use Cases
- Feature Requirements (Phase 1)
4.1. Platform & Security
4.2. Document Management
4.3. Capital Calls
4.4. Waterfall Calculations (Multi-Class)
4.5. Workflow Automation (Per Class)
4.6. Basic Analytics
4.7. AI Integration (Initial)
4.8. Architecture Design
4.9. Fund Accounting
4.10. Portfolio Tracking - Non-Functional Requirements
- Acceptance Criteria (Key)
- Success Metrics
- Risks & Mitigation
- Timeline (Sprint Breakdown)
- QA Documentation
Related Documents:
1. Product Overview¶
- Product Name: DealSphere
- Description: Secure, DLT-backed fund management platform for PE/VC with class-specific workflows and automated waterfalls, built on R3 Corda.
- Objective: Deliver an MVP that supports multi-class fund operations (permissions, capital calls, workflows, waterfalls), basic analytics, fund accounting, and portfolio tracking with AI-assisted experiences and a finalized architecture for scale.
2. Target Users¶
- General Partners (GP), Limited Partners (LP) by class (A, B, etc.)
- Fund Managers and Administrators
- Auditors and Compliance users
- Investment/Portfolio Analysts
- Investors (read-only class-specific views)
- AI Assistant (virtual role for automated tasks)
3. Core Use Cases¶
- Role-based, class-segregated access and views
- Document management with on-ledger metadata, versioning, and audit logs
- Capital call lifecycle per class (rules, notices, tracking)
- Multi-class European and American waterfalls with prioritization and clawbacks
- Class-specific workflow automation (approvals, reminders, escalations)
- Basic analytics by class (committed vs. deployed, portfolio breakdown)
- Fund accounting and NAV/P&L at class and combined levels
- Portfolio tracking with class-based contributions and returns
- AI-assisted class-specific queries and drafting
4. Feature Requirements (Phase 1)¶
4.1 Platform & Security¶
- Role-based access control (RBAC) with strict class segregation
- R3 Corda DLT integration for immutable audit trails
- Multi-tenant architecture supporting multiple funds
- Encryption at rest and in transit
- Compliance framework (GDPR, SOX, regional requirements)
4.2 Document Management¶
- Upload, version, and categorize documents per class
- On-ledger metadata storage with hash verification
- Access controls ensuring LP class segregation
- Document templates and automated generation
- Full audit logs for document lifecycle events
4.3 Capital Calls¶
- Class-specific capital call rules and thresholds
- Automated notice generation and distribution per class
- Payment tracking and reconciliation by class
- Default management and escalation workflows
- Integration with banking APIs for payment verification
4.4 Waterfall Calculations (Multi-Class)¶
- European and American waterfall models
- Class-based priority and allocation rules
- Clawback provisions and carry calculations
- Real-time distribution modeling and simulation
- Support for preferred returns and catch-up provisions
4.5 Workflow Automation (Per Class)¶
- Class-specific approval workflows
- Automated reminders and escalations
- Task assignment and tracking per class
- SLA monitoring and reporting
- Integration with email and notification systems
4.6 Basic Analytics¶
- Class-based committed vs. deployed capital tracking
- Portfolio performance breakdown by class
- Cash flow projections per class
- Basic reporting and export functionality
- Real-time dashboard views segregated by class
4.7 AI Integration (Initial)¶
- AI-assisted document drafting with class-specific parameters
- Intelligent data extraction from uploaded documents
- Class-aware query processing and responses
- Automated categorization and tagging
- Compliance checking and risk flagging
4.8 Architecture Design¶
- Microservices architecture with API-first approach
- Cloud-native deployment (AWS/Azure/GCP)
- Scalable data architecture supporting multi-class operations
- API gateway for integrations and external services
- Security model for strict class-based segregation and audit
4.9 Fund Accounting¶
- Multi-class general ledger
- NAV and P&L per class and combined
4.10 Portfolio Tracking¶
- Company profiles with investment history split by class
- Performance metrics per class and consolidated views
5. Non-Functional Requirements¶
- Mobile-responsive web app
- High security: DLT auditability, encryption, RBAC with class segregation
- GDPR-ready and regional compliance placeholders
- Cloud deployment (AWS/Azure/GCP)
- API-first for integrations and modular scale
6. Acceptance Criteria (Key)¶
- Permissions: LPs see only their class; adjustable without contract redeploy
- Documents: Versioning works; access logs accurate; hash verification consistent
- Capital Calls: Issued per class rules; LP payment statuses update correctly
- Waterfalls: Distributions per class match test vectors; switching preserves class logic
- Workflows: Class A and B flows run concurrently without conflict; reminders as configured
- Analytics: Reports filterable by class; exports work
- AI: Filters/outputs adhere to class constraints; drafts match class parameters
- Accounting: NAV computed separately by class and combined
- Portfolio: Class-specific contributions and returns displayed
7. Success Metrics¶
- Funds onboarded leveraging multi-class features
- Volume of class-specific capital calls and distributions
- Workflow automation throughput and SLA adherence
- Accuracy of waterfall outputs vs. expected vectors
- Frequency of AI-assisted actions
- Analytics/export usage by class
8. Risks & Mitigation¶
- Multi-class complexity: Test vectors and simulation harnesses
- Security/segregation: Pen testing, rigorous data access validation, on-ledger audit
- Scope creep: Phased deliveries with strict change control
- Compliance variance: Region placeholders and pilot validation
9. Timeline (Sprint Breakdown)¶
10. QA Documentation¶
Summary: Phase 1 delivers a secure, multi-class MVP across permissions, documents, capital calls, waterfalls, workflows, analytics, fund accounting, and portfolio tracking, with AI assist and an architecture ready for scale.