AstraAI : LLMs, Retrieval, and AST-Guided Assistance for HPC Codebases

Mahesh Natarajan, Xiaoye Li, Weiqun Zhang · 2026

Read AstraAI : LLMs, Retrieval, and AST-Guided Assistance for HPC Codebases by Mahesh Natarajan, Xiaoye Li, Weiqun Zhang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-27423.txt.

Description

Natarajan and co-authors present AstraAI, a coding aid for supercomputer software. The tool runs in a plain Linux terminal. It blends language models with search over the project code. It also reads syntax trees to map code structure. Those two sources build a rich prompt for each request. Edits stay scoped and match the host project style. Both local and cloud models can drive it. The team shows the tool at work on the AMReX framework.

Catalog record

CREATORMahesh Natarajan, Xiaoye Li, Weiqun Zhang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE22708
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.27423

Browse the full library · About the Tower of Babel