# MODIFIABLE REQUEST FOR PROPOSALS (RFP) TEMPLATE
## Project Title: Oracle 23ai Document AI & Intelligent OCR: Extract PDFs & Invoices into APEX & Vector Search
- **RFP Document Reference**: `VNA-RFP-020` (Companion to Solution Brief `VNA-SB-020`)
- **Issuing Organization**: `[INSERT ORGANIZATION / COMPANY NAME]`
- **Issue Date**: `[2026-10-27]`
- **Proposal Due Date**: `[INSERT DATE, e.g., +30 Calendar Days]`
- **Reference Architecture Portal**: https://oracle-23ai-document-ai-ocr.web.app/
- **Template Attribution**: This site is designed by Viscosity.AI

---

## SECTION 1: ISSUING ORGANIZATION PROFILE & PROCUREMENT SCHEDULE (MODIFIABLE)

> **Instructions for RFP Author**: Replace all bracketed placeholders (`[...]`) below with your organization's specific dates, contacts, and environment metrics before issuing to prospective bidders.

| Procurement Milestone | Target Date / Entry (Modifiable) |
|---|---|
| Issuing Organization Name | `[Insert Legal Entity Name]` |
| Primary Procurement / Technical Contact | `[Insert Name, Title, Email]` |
| RFP Issue Date | `[2026-10-27]` |
| Vendor Written Questions Due | `[Insert Date - Week 2]` |
| Organization Responses to Vendor Q&A | `[Insert Date - Week 3]` |
| Final Vendor Proposals & Pricing Due | `[Insert Date - Week 4]` |
| Technical Architecture Demos / Oral Presentations | `[Insert Date - Week 5]` |
| Target Contract Award & Phase 0 Kickoff | `[Insert Date - Week 6]` |

### 1.1 Executive Summary & Business Objectives
`[INSERT COMPANY NAME]` ("Organization") is soliciting competitive proposals from qualified enterprise architecture and engineering firms to modernize and transition our current **Unstructured PDFs, Scanned TIFFs, Invoices, MSAs/SOWs & Lab Reports** environment to **OCI Document Understanding, Multimodal LLMs, Oracle 23ai AI Vector Search & APEX**.

Our primary business and technical objectives are:
- Legacy template-based OCR tools break the moment a vendor changes their invoice layout or a multi-page contract contains nested tables and handwritten annotations.
- Connecting standalone third-party AI tools to enterprise ERPs requires stitching together separate vector databases (Pinecone/Weaviate), blob stores, Python glue servers, and custom review UIs.
- Regulated finance, healthcare, and legal workflows require deterministic confidence scoring and a fast side-by-side human verification screen whenever a field confidence score falls below threshold.

---

## SECTION 2: CURRENT ENVIRONMENT BASELINE INVENTORY (MODIFIABLE)

Bidders must size their technical delivery plan and fixed-fee milestone pricing against the baseline inventory table below:

| Inventory Dimension | Current Baseline Count / Footprint (Modifiable) | Target Modernization SLA / Objective |
|---|---|---|
| Legacy Source Stack (`Stack X`) | Unstructured PDFs, Scanned TIFFs, Invoices, MSAs/SOWs & Lab Reports | Full retirement / modernization |
| Target Architecture (`Stack Y`) | OCI Document Understanding, Multimodal LLMs, Oracle 23ai AI Vector Search & APEX | Production-grade HA/DR & security |
| Monthly PDFs, Invoices & Contracts Processed | `[Insert Count, e.g., 25000 pages/mo]` | 100% functional parity & optimization |
| Document Types & Extraction Schemas | `[Insert Count, e.g., 15 doc types]` | Automated AST & schema uplift |
| Target Cutover Downtime Window | `[e.g., < 15 Minutes / Near-Zero Downtime]` | Rehearsed parallel run + rollback plan |
| Security, Compliance & Identity | `[e.g., SSO/MFA via Entra ID / Okta, Row-Level VPD, Audit Logs]` | Zero critical/high CVEs at Go-Live |

---

## SECTION 3: MANDATORY TECHNICAL SCOPE OF WORK & DELIVERABLES (MODIFIABLE)

The selected partner shall execute the project across four gated phases (**Phase 0: AST Discovery & Architecture Blueprint**, **Phase 1: Core Foundation & Data/Code Conversion**, **Phase 2: Integration, Runtime Filtering & UAT**, and **Phase 3: Parallel Run, Cutover & Hypercare**).

### 3.1 Core Architectural Requirements
1. **[MODIFIABLE REQUIREMENT 1]**: Multi-Stage Ingestion & Layout-Aware OCR: Incoming documents from email inboxes, SFTP, or OCI Object Storage are processed through OCI Document Understanding / Veryfi / Nanonets and vision-language models using strict JSON Schema output enforcement.
2. **[MODIFIABLE REQUIREMENT 2]**: Unified Oracle 23ai Relational + JSON + AI Vector Storage: Instead of maintaining a separate vector database, extracted JSON payloads, `VECTOR` embeddings (`DBMS_VECTOR`), and relational ledger rows live inside a single ACID-compliant Oracle Database 23ai instance.
3. **[MODIFIABLE REQUIREMENT 3]**: Automated Entity Resolution & PO/Contract Matching: Extracted line items are automatically cross-checked against existing purchase orders, contract rate cards, and master vendor tables using hybrid SQL + Vector similarity queries.
4. **[MODIFIABLE REQUIREMENT 4]**: Human-in-the-Loop APEX 24.2 Verification Workbench: Documents with confidence scores below 95% or math discrepancies route to a side-by-side APEX split-screen viewer highlighting the exact bounding box on the PDF next to the editable form field.

### 3.2 Component-Level Conversion & Deliverable Matrix
Bidders must confirm compliance with each component conversion requirement below:

| Req ID | Legacy Artifact / Baseline Workload | Required Target Architecture | Engineering Method & Acceptance Criteria | Priority (Modifiable) |
|---|---|---|---|---|
| REQ-01 | Manual Email / PDF Invoice & Receipt Entry | Automated OCR Pipeline + 3-Way PO Match in Oracle 23ai | Straight-Through Processing + Bounding-Box Exception UI | `[Mandatory / Optional]` |
| REQ-02 | Unindexed PDF Contract / MSA / SOW Repositories | Structured Renewal/Rate Tables + Oracle 23ai AI Vector Search | Clause Extraction + In-Database RAG (`DBMS_VECTOR`) | `[Mandatory / Optional]` |
| REQ-03 | Siloed External Vector DBs & Python Glue Apps | Converged Oracle 23ai (Relational + JSON Duality + Vector) | Single Security, Backup & SQL Governance Plane | `[Mandatory / Optional]` |
| REQ-04 | Private Equity / Financial PDF Trial Balance Packs | Automated Medallion Lakehouse Ingestion & Validation | Deterministic Cross-Foot & Balance Tie-Out Rules | `[Mandatory / Optional]` |

### 3.3 Mandatory Platform & Tooling Standards
- Oracle Database 23ai (`VECTOR` data type, `DBMS_VECTOR`, JSON Relational Duality Views)
- OCI Document Understanding, OCI Generative AI (Cohere / Llama), and Oracle Functions `F(n)`
- Oracle APEX 24.2 Split-Screen PDF Bounding-Box Verification Workbench
- Oracle REST Data Services (ORDS) for webhook ingestion

---

## SECTION 4: VENDOR TECHNICAL Q&A QUESTIONNAIRE & SCORING RUBRIC

Bidders must provide direct, concise written responses to each question below. The Issuing Organization's evaluation committee will grade responses using the embedded benchmark criteria.

### Vendor Question 1: Why store vector embeddings inside Oracle Database 23ai instead of a dedicated vector database?
- **Bidder Response Requirement**: Provide a concrete architectural explanation, reference toolset, and verifiable production proof. Generic marketing statements will be scored zero.
- **Evaluation Benchmark (What High-Scoring Responses Must Demonstrate)**: In enterprise applications, you almost never run a pure vector search in isolation; you need to filter by customer ID, effective date, department security role, and contract status at the same time. Oracle 23ai executes hybrid relational + JSON + vector similarity joins in a single SQL query with zero data duplication.
- **Evaluator Score (0 to 10 Points)**: `[   / 10 ]`

### Vendor Question 2: How do you prevent LLM hallucinations on financial numbers and dates?
- **Bidder Response Requirement**: Provide a concrete architectural explanation, reference toolset, and verifiable production proof. Generic marketing statements will be scored zero.
- **Evaluation Benchmark (What High-Scoring Responses Must Demonstrate)**: Every extracted document passes through deterministic PL/SQL validation rules (line-item sum == invoice subtotal + tax == total, valid PO lookup, date range bounds). Any mathematical mismatch is flagged automatically in the APEX human-in-the-loop queue.
- **Evaluator Score (0 to 10 Points)**: `[   / 10 ]`

### Vendor Question 3: What is included in the downloadable PDF Solution Brief?
- **Bidder Response Requirement**: Provide a concrete architectural explanation, reference toolset, and verifiable production proof. Generic marketing statements will be scored zero.
- **Evaluation Benchmark (What High-Scoring Responses Must Demonstrate)**: The PDF Solution Brief includes the Oracle 23ai Converged Document AI Reference Architecture, sample `DBMS_VECTOR` hybrid SQL patterns, and our human-in-the-loop confidence routing matrix.
- **Evaluator Score (0 to 10 Points)**: `[   / 10 ]`

### Vendor Question 4: How do you prevent report and query fan-out bloat, and how do you guarantee zero vendor lock-in?
- **Bidder Response Requirement**: Detail your methodology for clustering redundant legacy queries/reports into canonical core SQL views with runtime or AI-inference filtering, and specify customer code/data ownership upon project completion.
- **Evaluation Benchmark (What High-Scoring Responses Must Demonstrate)**: Bidder must demonstrate deterministic AST SQL clustering (collapsing 75%-90% of cloned reports into core database views with runtime predicates), 100% customer ownership of all DDL, PL/SQL, application source code, and Terraform scripts, and zero proprietary runtime lock-in.
- **Evaluator Score (0 to 10 Points)**: `[   / 10 ]`

### Vendor Question 5: Describe your automated regression testing, parallel run, and zero-downtime cutover methodology.
- **Bidder Response Requirement**: Specify how data parity, business-rule equivalence, and rollback safety are verified prior to production cutover.
- **Evaluation Benchmark (What High-Scoring Responses Must Demonstrate)**: Bidder must provide automated row-count and cryptographic hash reconciliation, automated unit/UI test suites (e.g., utPLSQL, Playwright), and a rehearsed cutover runbook with sub-minute rollback capability.
- **Evaluator Score (0 to 10 Points)**: `[   / 10 ]`

---

## SECTION 5: COMMERCIAL PRICING & LICENSING MATRIX (MODIFIABLE)

Bidders must complete the fixed-price milestone table below and disclose all recurring software or cloud consumption assumptions.

| Milestone / Cost Category | Deliverable & Acceptance Gate | Fixed Fee / Annual Cost (USD) | Estimated Duration (Weeks) |
|---|---|---|---|
| **Phase 0: Discovery & AST Code/Schema X-Ray** | Verified inventory, dead-code manifest, core SQL fan-out map, and final blueprint | `$[Insert]` | `[2 - 3 Weeks]` |
| **Phase 1: Target Environment & Automated Uplift** | Provisioned target stack, schema migration, and core business logic conversion | `$[Insert]` | `[4 - 8 Weeks]` |
| **Phase 2: UI, Integrations, Reporting & UAT** | Completed application/portal screens, runtime reports, AOP templates, and UAT sign-off | `$[Insert]` | `[3 - 6 Weeks]` |
| **Phase 3: Production Cutover & Hypercare** | Rehearsed cutover, reconciliation sign-off, documentation, and 30-day hypercare | `$[Insert]` | `[2 - 3 Weeks]` |
| **Optional: Annual Managed Services / Support** | 24x7 monitoring, patching, DBA performance tuning, and quarterly enhancements | `$[Insert / yr]` | `[12 Months]` |

---

## SECTION 6: PROPOSAL EVALUATION SCORECARD (MODIFIABLE)

| Evaluation Criterion | Weight (%) | Description |
|---|---|---|
| **1. Technical Architecture & Automation Tooling** | `35%` | Depth of AST automation, query fan-out rationalization, database/cloud architecture, and zero lock-in. |
| **2. Verifiable Production Experience & References** | `25%` | Proven track record delivering identical `Oracle 23ai Document AI, OCR & Vector Search` engagements with measurable outcomes. |
| **3. Cutover Safety, Testing & Compliance** | `20%` | Rigor of data reconciliation, automated regression testing, security/compliance controls, and rollback safety. |
| **4. Commercial Value & Total Cost of Ownership (TCO)** | `20%` | Transparent fixed-fee milestone pricing and 3-to-5 year software/cloud licensing savings. |
| **TOTAL** | **`100%`** | |

---
*Template Reference: `VNA-RFP-020` | Companion Portal: https://oracle-23ai-document-ai-ocr.web.app/ | This site is designed by Viscosity.AI*