Engineering Tradeoffs

Architectural Decisions & Zero-Cloud Constraints

Every design choice in AXIOM OS prioritizes strict local privacy, minimal battery drain, and deterministic accountability over cloud convenience.

01 / Zero-Cloud Privacy Invariant
Core Invariant

Modern productivity agents rely heavily on cloud APIs to process context. This fundamentally compromises privacy when dealing with screen content, private messages, and personal work habits. AXIOM OS enforces a strict zero-cloud constraint.

Zero API Keys

No OpenAI, Anthropic, or external credentials required. The system works completely air-gapped.

Local Ollama Run

All intelligence executes directly on local CPU/GPU silicon using quantized GGUF weights.

Loopback Only

Telemetry never leaves the 127.0.0.1 interface, enforced with operating-system socket blocks.

02 / Local Model Selection (VRAM vs Latency)
Silicon Benchmarks

Running an LLM constantly in the background can monopolize system resources. We evaluated several open-weights models before settling on the 3-billion parameter class for everyday background execution.

Candidate ModelVRAM FootprintReasoning SpeedEngineering Verdict
Llama-3-8B~6.0 GBMediumHigh reasoning quality, but excessive memory drain on ultraportable laptops.
Qwen-2.5-3B (Selected)~2.5 GBSub-second (Very Fast)Optimal balance: Fits in unified memory with negligible thermal footprint.
Phi-3-Mini~2.2 GBFastFast inference, but exhibited context truncation on multi-hour timeline summaries.
03 / Determinism vs Hallucination Prevention
Ground Truth

A critical design decision was separating the factual measurement of behavior from the AI interpretation. The Go daemon writes strict facts to SQLite (such as "Chrome was open on youtube.com for 5 minutes"). The LLM is only given read-access to this data to form opinions.

This guarantees the AI cannot hallucinate past user actions; it can only judge recorded, immutable ground truth.