Low-Code Development Enhancement Integrating Large Language Models for Intelligent Code Assistance in Oracle APEX

Authors

  • Srikanth Reddy Keshireddy

DOI:

https://doi.org/10.51983/ijiss-2025.IJISS.15.2.46

Keywords:

Oracle APEX, Low-Code Development, Large Language Models (LLMs), Intelligent Code Assistance, Prompt- Based Code Generation

Abstract

The growing use of low-code platforms within the
business sphere has spurred the need for real-time, intelligent
code help tools that can effectively integrate citizen developers
with expert programmers. This research investigates the
application of Large Language Models (LLMs) into Oracle
APEX with the goal of improving developer productivity and
minimizing the cognitive effort required during task execution.
We created a novel system designed to operate within the Oracle
APEX environment that includes an LLM-powered code
suggestion engine. This system is able to respond to natural
language queries by providing context-aware code snippets in
PL/SQL, JavaScript, and SQL. Furthermore, it dynamically
responds to page items, session states, and application metadata
in real time.
Our architecture integrates Oracle APEX's RESTful interfaces
with external LLMs through a modular API orchestration layer
to enable effortless generation of accurate deployable code. An
extensive experimental evaluation with 48 developers from
different seniority levels was performed, including tasks from
UI sketching to business logic coding. Achieved results indicate
an average of 41 percent reduction in completion time and 34
percent reduction in manual code cut editing. Structured logical
scenarios achieved over 90 percent syntax accuracy, and users
reported high confidence in the system outputs. The analysis
also focuses on token consumption, domain-specific error
patterns, and code review log feedback loops.
The primary objective of this research study was to illustrate the
profound impacts information technologies, particularly LLMs,
had on the speed of achieving low-code application development
with Oracle APEX. In addition, the study intends to
demonstrate the set of strategies proposed suited for the
consistent implementation of intelligent assistants that aid in
elevating code quality while reducing barriers, enabling
enterprise-level application delivery with minimal manual effort
preconfigured.

Downloads

Published

25-06-2025

How to Cite

Keshireddy, S. R. (2025). Low-Code Development Enhancement Integrating Large Language Models for Intelligent Code Assistance in Oracle APEX. Indian Journal of Information Sources and Services, 15(2), 380–390. https://doi.org/10.51983/ijiss-2025.IJISS.15.2.46