Technology.
Strategy.
Intelligence.
Sebastian Maniak on running AI agents in production — agentgateway, kagent, MCP, A2A, and the Kubernetes plumbing that keeps them fast, observable, and safe.
GitHub MCP Token Economics: Why Search Mode Cuts Your LLM Bill by ~60%
Every time an LLM talks to an MCP server, it has to be told what tools exist. That tool catalog — the JSON schema of every tool, its parameters, and descriptions — is injected into the prompt on every single turn. For a …
Read the article →Latest writing
View the full archive →Running agentgateway on Proxmox (LXC): the Everything MCP server end-to-end
Running agentgateway on Proxmox (LXC): the Everything MCP server end-to-end Date: June 2026 Author: Sebastian Maniak Tags: agentgateway, mcp, proxmox, lxc, …
One-Script Deployment: agentgateway + Self-Hosted Langfuse on Kubernetes for LLM Cost Analysis
Running production-grade LLM gateways with full observability usually involves many manual steps across Helm charts, CRDs, tracing configuration, and UI key …
agentgateway Standalone → Langfuse: Direct OTLP Tracing (No Collector)
I’ve already covered the production OTel Collector pattern for shipping agentgateway traces to Langfuse. This is the opposite end of the spectrum: the …
Langfuse Integration with agentgateway (OTel Collector Pattern) for cost controls, observability
agentgateway emits rich OpenTelemetry traces for every LLM request, tool call, and policy decision. This guide shows the production-grade way to forward those …
Self-Hosted Langfuse with Docker + kagent Integration
Running Langfuse as a self-hosted Docker stack gives you full control over your LLM observability data. This guide shows how to deploy it and integrate it with …
Running kagent on a kind Cluster with a Local vLLM + Qwen3 Backend
Running kagent on a kind Cluster with a Local vLLM + Qwen3 Backend By Sebastian Maniak Most kagent walkthroughs assume you’re pointing at OpenAI or …