Pythia: AI-assisted Code Completion System

TL;DR

Pythia is an AI-powered code completion system that leverages deep learning models trained on code syntax to provide fast, accurate method and API suggestions integrated into Visual Studio Code.

Contribution

It introduces a novel end-to-end deep learning approach for code completion that outperforms traditional models in accuracy and integrates seamlessly into IDEs.

Findings

Top-5 accuracy of 92% on Python repositories

Surpasses baseline models by 20% in accuracy

Predicts code completions within 100 ms

Abstract

In this paper, we propose a novel end-to-end approach for AI-assisted code completion called Pythia. It generates ranked lists of method and API recommendations which can be used by software developers at edit time. The system is currently deployed as part of Intellicode extension in Visual Studio Code IDE. Pythia exploits state-of-the-art large-scale deep learning models trained on code contexts extracted from abstract syntax trees. It is designed to work at a high throughput predicting the best matching code completions on the order of 100 $ms$. We describe the architecture of the system, perform comparisons to frequency-based approach and invocation-based Markov Chain language model, and discuss challenges serving Pythia models on lightweight client devices. The offline evaluation results obtained on 2700 Python open source software GitHub repositories show a top-5 accuracy of 92\%, surpassing the baseline models by 20\% averaged over classes, for both intra and cross-project settings.