LLM Security: Hardening Tool-Use Architectures Against Data Leakage
Moving beyond prompt engineering to secure agentic systems via strict schema constraints, input sanitization, and deterministic function-calling control flows.
31 July 2026
AI agent builder at a Sapporo robotics company, writing tool-use and function-calling integrations for warehouse automation. Designed the multi-step planning loop that achieves 95% task completion without human intervention.
Moving beyond prompt engineering to secure agentic systems via strict schema constraints, input sanitization, and deterministic function-calling control flows.
31 July 2026
This article delves into advanced state management techniques using Redux in Flutter applications. It explores the Redux architecture, the implementation of actions and reducers, middleware integration, and the optimization of state changes for efficient rendering. We will also discuss selective state updates and the effects of concurrent operations on performance, providing practical code snippets for real-world applications.
23 July 2026
A practical guide to designing multi-agent systems where specialised agents hand off tasks to each other, share memory, and coordinate without losing context across long-running workflows
15 July 2026