Local Experiments
Notes from hands-on trials, benchmarks, and real-world setup work.
Soothsayer MCP kernel: from prompts to controlled orchestration
How I built a policy-governed MCP runtime where models can reason freely but execution stays deterministic, verifiable, and auditable.
Ollama launch: integration handoff
A new launcher flow that hands local models into tools like Codex and OpenCode.
AWS GPU instance baseline
A minimal cloud baseline focused on identity, networking, and cost control.
GCP GPU instance baseline
A first cloud baseline with focus on projects, IAM, and network rules.
Ollama setup: local baseline
A first clean install that taught me where local stacks really fail.
Azure setup: first cloud pass
A first-time Azure run that turned identity and networking into the real curriculum.
Ollama on M1: thermal baseline
A baseline run to learn where M1 throttles and why it matters.
Context window stress test
A small experiment to see where longer context starts to degrade quality.