Oracle AI-Driven Modernization Developer

Job Description:
Introduction:

• This Statement of Work (SOaW) outlines the tasks, deliverables, and responsibilities associated with the AI-driven modernization of legacy Oracle Forms and Oracle Framework applications to Oracle APEX for client.
• The goal is to enhance operational efficiency and reduce costs and application delivery timelines through the application of AI technologies.

Background:

• Client currently utilizes customized legacy Oracle Forms and Oracle Framework applications for various business processes. As part of an effort to modernize our portfolio, Client would like to transition these applications to Oracle APEX. This project aims to leverage AI to expedite the delivery of APEX applications.

Scope of Work:
The scope of work includes the following key activities:

• AI Model Training & Fine-Tuning: Utilize open-source, air-gapped large language models (LLMs) for secure processing. Train and fine-tune models using legacy application exports (Oracle Forms and Oracle Framework applications) to extract business rules, logic, and metadata. Implement an iterative training process to improve model accuracy from an initial 50-60% to 80-90%.
• Automated Code & Rule Extraction: Parse and clean legacy code (Oracle Forms, PL/SQL) to extract key business rules, triggers, procedures, and configurations. Deliver outputs in standardized formats (PL/SQL scripts, JSON boilerplates) for use in Oracle APEX.
• Accelerated APEX Application Modernization: Rapidly generate APEX application boilerplates from extracted logic. Provide downloadable scripts for database object creation, reducing manual recoding efforts.
• Security & Best Practices: Ensure all processing occurs within secure environments, with no proprietary data shared externally. Enforce best practices in code structure, documentation, and naming conventions. Review Client best practices in APEX code structure, documentation, and naming conventions and update output to align to Client standards where logical.
• Scalability & Reusability: Extend the AI-driven approach to other legacy modules or applications as needed, supporting future automation opportunities.

Mandatory Requirements:

• The assigned Subcontractor personnel shall have a bachelor's degree from an accredited college or university with major coursework in Computer Science, Management of Information Systems, or related field or equivalent experience.
• The Subcontractor personnel shall have at least five years of experience with Oracle APEX development, and legacy system migration. Resumes or qualifications should be submitted for review.
• The assigned Subcontractor personnel shall have relevant experience in AI-driven modernization projects using legacy Oracle Forms OR Oracle Framework applications. This includes providing case studies or references from similar projects.

Tasks:

• Conduct a kickoff meeting with Client stakeholders to align on project objectives, timelines, and communication protocols.
• Develop and train AI models using legacy application data to extract business rules, logic, and metadata relevant to the modernization effort.
• Automate the extraction of key business rules, triggers, procedures, and configurations from the legacy systems, delivering outputs in standardized formats.

• Design and develop APEX applications based on the extracted logic and requirements, ensuring alignment with Client's standards and best practices.
• Create comprehensive documentation for the modernization process, including design specifications, code documentation, user manuals, and training materials.
• Conduct training sessions for Client personnel to facilitate knowledge transfer regarding the AI models and modernization process.
• Manage any changes to the project scope, timelines, or resources through a formal change management process, ensuring that all changes are documented and approved.
• Schedule and conduct regular progress check-ins with Client stakeholders to provide updates on project status, address any issues, and ensure alignment with project goals.
• Conduct a final project review meeting to present the completed work, gather feedback, and discuss any potential next steps.

Deliverables:

• AI models trained and fine-tuned for Client's specific needs.
• Extracted business rules and legacy code in standardized formats.
• APEX application boilerplates and scripts for database object creation.
• Documentation detailing the AI models developed, including training methodologies, data sources, model performance metrics, and any assumptions made during the modeling process.

Scope Options:
Pending successful completion of the initial modernization project, Client National Laboratories intends to explore the opportunity to extend the AI-driven modernization approach to additional legacy applications. This may include, but is not limited to, the following:

• Assessment of Additional Legacy Applications: Conduct assessments of other legacy Oracle Forms and Oracle Framework applications to identify modernization opportunities and prioritize them based on business needs.
• Phased Modernization Approach: Implement a phased approach to modernizing multiple legacy applications, leveraging the methodologies and best practices established during the initial project to ensure efficiency and consistency.
• Scalability of AI Solutions: Utilize the AI models and extraction techniques developed in the initial project to facilitate the rapid modernization of additional applications, aiming to replicate the success achieved in the first phase.
• Resource Allocation: Allocate resources and establish timelines for subsequent projects based on the lessons learned and successes from the initial modernization effort.
• Continuous Improvement: Incorporate feedback and insights gained from the initial project to refine processes and enhance the effectiveness of future modernization initiatives.

Assumptions:

• All work will be conducted remotely without access to Client technology or cyber accounts.
• NTESS will provide access to relevant functional, technical, and business resources.