Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{189583,
        author = {Nishmitha Shetty and Harshitha L and Sathvik P and MD Asif and Dr.Mamatha M},
        title = {Code Whisperer: A Local AI-Powered Code Generation and Execution Platform},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {7},
        pages = {6815-6829},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=189583},
        abstract = {Code Whisperer is a locally deployed AI-assisted programming system that converts natural-language problem descriptions into executable source code across multiple programming languages. The system utilizes a pre-trained large language model, LLaMA 3.2, executed locally through the Ollama runtime, replacing earlier cloud-based and API-dependent code generation models commonly used in existing tools. Unlike prior approaches that rely on remote inference, the proposed system operates entirely offline, ensuring data privacy, reduced latency variability, and transparent execution control. The platform is implemented using a React-based frontend and a Fast API backend, which collaboratively manage prompt handling, code normalization and validation, secure execution, and result visualization. Experimental evaluation demonstrates an average functional correctness of 85.4% across Python, C/C++, Java, and JavaScript tasks. However, the system is constrained by local hardware limitations and reduced model capacity compared to large-scale cloud-hosted solutions. Despite these limitations, the results show that effective coordination between local AI inference and controlled execution mechanisms can deliver reliable and practical code generation in a fully offline environment.},
        keywords = {Local Code Generation, Large Language Models, LLaMA 3.2, Offline AI Systems, React Frontend, Fast API Backend, Secure Code Execution, Software Engineering Tools.},
        month = {December},
        }