AI Agents
Embeddings as Database Indexes: Tool-Call Permissions in AI Agents
How tool-call permissions and approval boundaries shape reliable AI agent systems using embeddings as the new database index
Blog
Thoughts on AI engineering, backend, frontend, and building modern software.
AI Agents
Why Multi-Agent Systems Fail: Tool-Call Permissions and Approval Boundaries
Most multi-agent systems fail in production due to uncontrolled tool-call permissions. Learn how to avoid catastrophic errors with strict approval boundaries and state-aware design.
AI Agents
Securing AI Agent Tool Access: Avoiding Production Pitfalls
Learn how to secure AI agent tool access, prevent unauthorized actions, and build robust production systems with clear permissions and observability.
AI Agents
Building Production-Grade AI Agents: Lessons from the Trenches
Learn real-world strategies for building reliable AI agents in production, including tradeoffs, failure modes, and design patterns from a senior engineer's perspective.