# PopPy: Opportunistically Exploiting Parallelism in Python Agentic Workflows

arXiv: https://arxiv.org/abs/2605.18697 · HTML: https://mlsanity.com/paper/2605.18697/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 74 references that are in the map.
- First posted on arXiv 18 May 2026.

## Builds on (most cited first, 12 of 12)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [ViperGPT: Visual Inference via Python Execution for Reasoning](https://mlsanity.com/paper/2303.08128/) (2023)
- [Executable Code Actions Elicit Better LLM Agents](https://mlsanity.com/paper/2402.01030/) (2024)
- [Skeleton-of-Thought: Prompting LLMs for Efficient Parallel Generation](https://mlsanity.com/paper/2307.15337/) (2023)
- [Rethinking LLM Uncertainty: A Multi-Agent Approach to Estimating Black-Box Model Uncertainty](https://mlsanity.com/paper/2412.09572/) (2024)
- [Opportunistically Parallel Lambda Calculus](https://mlsanity.com/paper/2405.11361/) (2024)
- [PaSh: Light-touch Data-Parallel Shell Processing](https://mlsanity.com/paper/2007.09436/) (2020)
- [Dias: Dynamic Rewriting of Pandas Code](https://mlsanity.com/paper/2303.16146/) (2023)
- [Assessing Large Language Models in Comprehending and Verifying Concurrent Programs across Memory Models](https://mlsanity.com/paper/2501.14326/) (2025)
- [Identifying Concurrency Bug Reports via Linguistic Patterns](https://mlsanity.com/paper/2601.16338/) (2026)
- [Embodied Tree of Thoughts: Deliberate Manipulation Planning with Embodied World Model](https://mlsanity.com/paper/2512.08188/) (2025)

## Abstract (arXiv)

Agentic workflows, which compose calls to ML models using a general-purpose programming language like Python, are widely used for a variety of user-facing tasks, from software engineering to enterprise automation, making their end-to-end latency a critical bottleneck. To improve their latency, developers can either manually parallelize them or use restricted agentic frameworks that cannot express or exploit all forms of parallelization. We address this by developing PopPy, a system that parallelizes external calls, e.g., AI models, in agentic workflows written in a very expressive subset of Python. PopPy combines an ahead-of-time compiler with a runtime, addressing three key challenges in extracting parallelism from Python applications: language complexity, dynamic dispatch, and variable mutation. On a set of real-world agentic workflows, PopPy achieves up to 6.7× speedups in end-to-end execution time, compared to standard Python execution and several agentic frameworks, while preserving the sequential program semantics.