Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

AISAC: An Integrated multi-agent System for Transparent, Retrieval-Grounded Scientific Assistance

Chandrachur Bhattacharya, Sibendu Som

TL;DR

The paper tackles the need for grounded, memory-rich, and transparent AI in scientific workflows. It introduces AISAC, an integrated multi-agent system that stacks LangGraph orchestration, hybrid memory (SQLite + FAISS), and a retrieval-grounded architecture with incremental RAG indexing, all wrapped in a declarative project bootstrap layer. Key contributions include a three-axis extensibility model, a graph-based Router–Planner–Coordinator reasoning workflow with dedicated helper agents, and a live graphical interface that provides end-to-end provenance and traceability. Demonstrations across Argonne domains show AISAC’s cross-domain applicability, reproducibility, and transparent decision-making, aligning with DOE and NSF priorities for trustworthy AI in science. Overall, AISAC presents a reusable foundation for domain-specific AI co-scientists that maintain methodological consistency while enabling domain adaptation and auditable scientific reasoning.

Abstract

AI Scientific Assistant Core (AISAC) is an integrated multi-agent system developed at Argonne National Laboratory for scientific and engineering workflows. AISAC builds on established technologies - LangGraph for orchestration, FAISS for vector search, and SQLite for persistence - and integrates them into a unified system prototype focused on transparency, provenance tracking, and scientific adaptability. The system implements a Router-Planner-Coordinator workflow and an optional Evaluator role, using prompt-engineered agents coordinated via LangGraph's StateGraph and supported by helper agents such as a Researcher. Each role is defined through custom system prompts that enforce structured JSON outputs. A hybrid memory approach (FAISS + SQLite) enables both semantic retrieval and structured conversation history. An incremental indexing strategy based on file hashing minimizes redundant re-embedding when scientific corpora evolve. A configuration-driven project bootstrap layer allows research teams to customize tools, prompts, and data sources without modifying core code. All agent decisions, tool invocations, and retrievals are logged and visualized through a custom Gradio interface, providing step-by-step transparency for each reasoning episode. The authors have applied AISAC to multiple research areas at Argonne, including specialized deployments for waste-to-products research and energy process safety, as well as general-purpose scientific assistance, demonstrating its cross-domain applicability.

AISAC: An Integrated multi-agent System for Transparent, Retrieval-Grounded Scientific Assistance

TL;DR

The paper tackles the need for grounded, memory-rich, and transparent AI in scientific workflows. It introduces AISAC, an integrated multi-agent system that stacks LangGraph orchestration, hybrid memory (SQLite + FAISS), and a retrieval-grounded architecture with incremental RAG indexing, all wrapped in a declarative project bootstrap layer. Key contributions include a three-axis extensibility model, a graph-based Router–Planner–Coordinator reasoning workflow with dedicated helper agents, and a live graphical interface that provides end-to-end provenance and traceability. Demonstrations across Argonne domains show AISAC’s cross-domain applicability, reproducibility, and transparent decision-making, aligning with DOE and NSF priorities for trustworthy AI in science. Overall, AISAC presents a reusable foundation for domain-specific AI co-scientists that maintain methodological consistency while enabling domain adaptation and auditable scientific reasoning.

Abstract

AI Scientific Assistant Core (AISAC) is an integrated multi-agent system developed at Argonne National Laboratory for scientific and engineering workflows. AISAC builds on established technologies - LangGraph for orchestration, FAISS for vector search, and SQLite for persistence - and integrates them into a unified system prototype focused on transparency, provenance tracking, and scientific adaptability. The system implements a Router-Planner-Coordinator workflow and an optional Evaluator role, using prompt-engineered agents coordinated via LangGraph's StateGraph and supported by helper agents such as a Researcher. Each role is defined through custom system prompts that enforce structured JSON outputs. A hybrid memory approach (FAISS + SQLite) enables both semantic retrieval and structured conversation history. An incremental indexing strategy based on file hashing minimizes redundant re-embedding when scientific corpora evolve. A configuration-driven project bootstrap layer allows research teams to customize tools, prompts, and data sources without modifying core code. All agent decisions, tool invocations, and retrievals are logged and visualized through a custom Gradio interface, providing step-by-step transparency for each reasoning episode. The authors have applied AISAC to multiple research areas at Argonne, including specialized deployments for waste-to-products research and energy process safety, as well as general-purpose scientific assistance, demonstrating its cross-domain applicability.

Paper Structure

This paper contains 43 sections, 1 figure, 1 table.

Table of Contents

  1. Introduction
  2. AISAC's Extensibility Model:
  3. System Architecture
  4. 1. Orchestration Layer.
  5. 2. Memory and Persistence Layer.
  6. 3. Retrieval and Knowledge Integration.
  7. 4. Tooling and Action Layer.
  8. 5. Configuration and Extensibility Layer.
  9. 6. Graphical Transparency Interface.
  10. 7. Agent Prompt and Personality Registry.
  11. Multi-agent Roles
  12. Memory and Retrieval Layer
  13. Tooling and Action Space
  14. 1. Information Retrieval and Literature Search.
  15. 2. Local Semantic Retrieval.
  16. ...and 28 more sections

Figures (1)

  • Figure 1: High-level architecture of the AI Scientific Assistant Core (AISAC).