Publication Date: 8/25/2026

Event: The 3rd Workshop on High-Performance eScience (HiPES 2026) in conjunction with Euro-Par 2026

Reference: pp. 1-12, 2026

Authors: Giuseppe Coviello, NEC Laboratories America, Inc.; Mohammad A. Khojastepour, NEC Laboratories America, Inc.; Kunal Rao, NEC Laboratories America, Inc.; Willard Dennis, NEC Laboratories America, Inc.; Srimat T. Chakradhar, NEC Laboratories America, Inc.

Abstract: Workflows that use generative AI are error-prone and hard to audit when model calls, validation, retries, and outputs are scattered across prompts, scripts, and configuration. AI native workflows need these elements connected in one place so the pipeline can be audited and reviewed end to end. This paper presents Völ, a domain-specific language for expressing AI native tasks, including scientific workflows. Völ separates deterministic control flow from probabilistic model calls and places validation, retry logic, and failure handling next to the task that uses them, so the policy for each model step is visible where it runs. It expresses model calls, structured-output contracts, automatic repair, retry policy, and workflow control in one program rather than across prompts, helper scripts, libraries, and glue code, so end-to-end behavior can be read and reviewed in one place. We use chemical entity recognition in scientific articles from NLM-Chem to evaluate Völ and compare it with Python, LangGraph, Snakemake, and Nextflow implementations. The Völ program is more compact and achieves the lowest mean end-to-end runtime, including about 1.8× and 2.1× speedups over Python and LangGraph at 16 concurrent threads. On a six-node SLURM cluster, Nextflow with embedded Völ runs about 1.5× faster than Nextflow with embedded Python and keeps per-job AI logic more compact than embedding Python in Nextflow or Snakemake.

Publication Link: