How We Route Cloud APIs, Hermes, OpenClaw, and Your Mac
Hermes and OpenClaw run on your machine or VM. They can call cloud APIs or local Ollama; AICoven does not choose. We route HTTP to your agent and WebSocket to your Mac.
Engineering Blog
Opinionated field notes on agent architecture, BYOK orchestration, local LLMs, and prompt economics — from building AICoven on iPhone, iPad, and Mac.
How We Rebuilt Tool Calling: From 75 Schemas to Smart Selection
Anthropic's advanced tool use research showed us the path. We adapted it for every provider — and for local models that can't even call tools natively.
The MCP Token Tax: How to Manage 200+ Tools on an 8K Budget
Dumping 200 MCP tool schemas into a local model's context is a great way to ensure it never answers your question. Here is how we fixed the 'amnesiac expert' problem.
Deterministic UI in the Age of Streaming AI (And Why It Keeps Breaking)
AI doesn't return clean JSON — it streams chaotic fragments of text, tool calls, and errors in real time. Here's what it took to render that smoothly, and why it still breaks.
The Context Cost: Why Context Management is the Hardest Problem in Local AI
When you build a local-first AI app, you hit a brutal hardware constraint almost immediately: the context window. Every token is precious. Here's how we're managing it.
The Reality of 'Local Agents': API Tools vs. Prompt Engineering
Building native cloud APIs is easy. Getting a local Llama-3 model to execute shell commands requires the dark arts of prompt engineering and a lot of regex.
The Context Sandwich: Why 'Infinite Memory' is a Trap
Everyone wants an infinite context window. The math says no. Here's how we compress history to keep the system prompt from getting lost in the middle.
The JSON Mode Tax: Why Your Agent Bills Are Bloated
We love JSON mode because it makes agents reliable. But forcing a model to think in JSON burns 3x more tokens than letting it reason in plain text.
The Middleman Attack: Why Your AI Wrapper is Logging Your Code
If you aren't using your own API key, you're trusting a middleman with your codebase. Here's how the big three handle your data differently depending on how you access them.