# Insights from the Glass Box

Field notes on agentic AI, the Model Context Protocol, observability, and building transparent, portable AI workflows.

### Engineering Articles

- [**MCP Server as a Service: Your Data, Behind Your Firewall**](/content/site/ob/mcp-server-as-a-service/index.html)  
  Every AI client now speaks MCP. The question is whose server exposes your business data — and whether it runs behind your firewall, governed and battle-tested.

- [**OBTO vs Lovable & Base44: Vibe Coding You Can Actually Own**](/content/site/ob/obto-vs-lovable-base44/index.html)  
  Where the AI app builders shine, where they keep your software inside their walls, and what "Describe it. Ship it. Own it." actually means.

- [**OBTO vs LangSmith vs Langfuse: Choosing a Self-Hosted AI Stack**](/content/site/ob/obto-vs-langsmith-vs-langfuse/index.html)  
  Where the three tools actually overlap, what self-hosting really costs for each, and an honest guide to choosing your agent stack.

- [**LLM Data Cleaning: Patterns for Production ETL Pipelines**](/content/site/ob/llm-data-cleaning/index.html)  
  Rules first, model second, humans last — hybrid cleaning pipelines, entity resolution without O(n²) costs, validation gates, and cost control.

- [**AI Ticket Triage: How Agents Automate the Helpdesk**](/content/site/ob/ai-ticket-triage/index.html)  
  Where triage agents actually work, the guardrails they need, and the three metrics that prove helpdesk automation is paying off.

- [**AI Agent Cost Tracking: Know What Every Run Costs**](/content/site/ob/ai-agent-cost-tracking/index.html)  
  Why agent loops make token math non-linear, how to compute cost per completed task, and the five cost levers ranked by impact.

- [**Multi-Model Orchestration: Routing LLMs by Cost and Task**](/content/site/ob/multi-model-orchestration/index.html)  
  Route each request by cost, latency, and task instead of paying frontier prices for everything — the strategies that work, and the hidden costs.

- [**How to Host a Remote MCP Server in 2026: A Practical Guide**](/content/site/ob/host-remote-mcp-server/index.html)  
  Transport, OAuth 2.1, scaling, and observability — what it actually takes to keep a remote MCP server alive in production.

- [**How to Build an MCP Tool: A Practical Guide**](/content/site/ob/build-an-mcp-tool/index.html)  
  The three MCP primitives, stdio vs. Streamable HTTP transport, a worked example, and what it takes to run a tool in production.

- [**AI Agent Observability: A Practical Guide for Teams**](/content/site/ob/ai-agent-observability/index.html)  
  What observability actually requires for agents — traces, tool calls, per-step token cost, and replay — and how to instrument it.

### Vision Articles

- [**How OBTO is Defining the AI Workforce of the Future**](/content/site/ob/ai-workforce/index.html)  
  How transparent, observable agents are reshaping the way teams build and deploy an AI workforce.

- [**Support for Groq: Fast Inferencing for Better AI Workflows**](/content/site/ob/fast-ai-inference/index.html)  
  Bringing Groq's high-speed inference to OBTO to make agentic workflows faster and more cost-efficient.

### Architecture

## Design & Architecture  
  A visual walkthrough of the OBTO Glass Box architecture — how the runtime, observability, and MCP layers fit together.