Conversational AI workspace improves memory and context management for developers
An Interactive GraphRAG Workspace for Reifying, Auditing, and Steering Conversational AI Memories
Software Engineering
Summary
Developers using chat-based AI tools for coding and software design often face problems when trying to continue work across multiple sessions because the AI’s memory is unstructured and becomes confusing over time. The authors created a system called CognitiveCanvas that turns conversation logs into organized, interactive graphs, making it easier to track goals, facts, and code. Their tool also checks for outdated information and helps users fix errors with a single click. This approach reduces clutter and helps programmers keep conversations more useful and accurate.
What this means in practice
- •For software development teams: Improve management and reuse of AI conversation context across multiple coding sessions by structuring dialogue as interactive knowledge graphs.
- •For technical support teams: Use graph-based memory tools to audit and update technical conversation histories, reducing errors caused by outdated software knowledge.
Authors
Munshi Md Arafat Hussain, Md. Abdur Rahman, Md Saeed Siddik
Abstract
Conversational Large Language Models (LLMs) are widely employed in software engineering for code generation, debugging, and architectural design. However, as developers engage in multi-session development tasks, technical knowledge becomes siloed in isolated, linear chat histories. Reusing context across sessions leads to three critical challenges: (1) flat-vector retrieval fails to traverse multi-hop relational dependencies between distant architectural decisions; (2) temporal drift in software packages and APIs results in silent context poisoning when reusing aged dialogue logs; and (3) raw conversation concatenation induces prompt bloat and attention degradation. We present CognitiveCanvas, an interactive tool that transforms conversational AI memories from ephemeral text logs into persistent, manipulable, and audited knowledge objects. \textsc{CognitiveCanvas} integrates: (1) a Hierarchical Semantic Fact-Graph (HSFG) that distills dialogues into Goal, Fact, and Code artifacts, achieving a 92.2\% token reduction while enabling multi-hop GraphRAG traversal; (2) a Continuous Temporal Drift Auditor coupling 60+ regex heuristics with live calendar-anchored LLM verification for 1-click self-correction; and (3) a Spatial-Symbolic Prompt Compiler (SSPC) that translates 2D direct manipulation gestures (Voronoi clustering, node resizing, synthesis edges) directly into structured attention directives