Notes on building AI-agent platforms, backend systems, and internal developer tools — by Siva Prasad Raju Bandi.https://musings.sivaapps.com/https://musings.sivaapps.com/blog/building-an-agent-orchestrator/https://musings.sivaapps.com/blog/building-an-agent-orchestrator/A hand-rolled state machine for LLM-driven chat: intent routing, tool invocation, SSE streaming, and out-of-band completion — and why we skipped LangGraph.Tue, 25 Nov 2025 00:00:00 GMThttps://musings.sivaapps.com/blog/designing-an-intent-classification-service/https://musings.sivaapps.com/blog/designing-an-intent-classification-service/The front-of-funnel decision — direct answer, clarify, or call a tool — and why building it as a small, dedicated service pays off.Thu, 20 Nov 2025 00:00:00 GMThttps://musings.sivaapps.com/blog/durable-log-plus-event-bus/https://musings.sivaapps.com/blog/durable-log-plus-event-bus/One service becomes the audit trail and the dispatcher. How persistence-before-fanout, HMAC replay protection, and presigned-URL attachment sync fit together.Sun, 30 Nov 2025 00:00:00 GMThttps://musings.sivaapps.com/blog/n8n-as-an-agent-runtime/https://musings.sivaapps.com/blog/n8n-as-an-agent-runtime/Using n8n as the "tool provider" for an LLM agent. Sub-workflows become tools the model can call. Changing agent capabilities becomes a workflow edit, not a deploy.Wed, 10 Dec 2025 00:00:00 GMThttps://musings.sivaapps.com/blog/two-mcp-server-patterns/https://musings.sivaapps.com/blog/two-mcp-server-patterns/Building Model Context Protocol servers for two very different jobs — fronting a SaaS with an OAuth broker, and exposing an internal knowledge base to authenticated users.Fri, 05 Dec 2025 00:00:00 GMT