Agint: Agentic Graph Compilation for Software Engineering Agents
Abhi Chivukula, Jay Somasundaram, Vijay Somasundaram
TL;DR
Agint tackles the core limitations of LLM-based coding agents by introducing an agentic graph compiler that converts natural-language intent into typed, effect-aware directed acyclic graphs using a six-floor type system $TEXT \to TYPED \to SPEC \to STUB \to SHIM \to PURE$. The system combines a hierarchical, locality-aware compilation pipeline with a hybrid runtime and a composable Unix-style toolchain, enabling parallelized, reproducible development and execution. Key contributions include the type-floor progression, locality-preserving compilation, and an effect-aware, JIT-like runtime supported by Flyte/Hydantic orchestration and Schemagin/Datagin data tooling. By enabling early execution, speculative evaluation, and cross-team collaboration across code and data, Agint promises faster prototyping, more reliable production pipelines, and scalable agentic workflows, while acknowledging dependencies on LM quality and domain coverage.
Abstract
LLM-based coding agents are increasingly common but still face challenges in context management, latency, reliability, reproducibility, and scalability. We present Agint, an agentic graph compiler, interpreter, and runtime that incrementally and hierarchically converts natural-language instructions into typed, effect-aware code DAGs. Agint introduces explicit type floors (text to data to spec to code) grounded in semantic graph transformations and a hybrid LLM and function-based JIT runtime. This enables dynamic graph refinement, reproducible and optimizable execution, speculative evaluation, and interoperability with existing developer tools. Agint's typed graph bindings improve reliability and allow concurrent composition of concurrent codebases by construction, supporting accelerated development with smaller and faster models, lower latency, efficient context utilization, and higher throughput. Hierarchical compilation allows scalable graph edits, while the graph structure supports reproducibility and efficient parallel generation. Agint provides a composable unix-style toolchain: dagify (DAG compiler), dagent (hybrid JIT runtime), schemagin (schema generator), and datagin (data transformer) for realtime, low-latency code and dataflow creation. Human developers and coding agents refine graphs through the Agint CLI, while non-technical users use Agint Flow GUI for visual editing, conversational refinement, and debugging to promote prototype agentic workflows to production code. This continuous co-creation model allows teams to prototype quickly, refine seamlessly, and deploy reliably, bridging natural language, compiler methods, and developer tooling to enable a new generation of composable, team-centric coding agents at scale.