DocsGetting Started

Getting Started

Your AI identity in 20 minutes

A complete walkthrough from account creation to your first API call — understand how the 4-layer blueprint system captures who you are.

~20 min read

On this page

Quick walkthrough Blueprint layers The interview First API call Next steps

Quick walkthrough

Four steps to full identity coverage

Artificial ID is a structured pipeline: create, interview, build, distribute. Each step feeds the next, and the whole thing takes less than half an hour the first time.

Create your account

Sign up with Google OAuth, a passkey, or email. Your account is the container — the blueprint stays encrypted in your own Google Drive.

Takes about 60 seconds. No credit card required for the free tier.

Complete the AI interview

A guided, adaptive interview across 9 phases captures how you think, communicate, code, and work. One question at a time — no long forms.

Typically 15–20 minutes. You can pause and resume anytime.

Review your blueprint

After the interview, your AI behavioral blueprint is assembled across 4 layers and 10+ modules. Review and edit individual modules at any time.

The blueprint is encrypted client-side before it ever leaves your browser.

Distribute your identity

Push your blueprint to any AI tool: connect via MCP, install the browser extension, call the REST API, or export as markdown.

All 5 distribution channels are available on the Pro plan.

Blueprint structure

What is an AI behavioral blueprint?

Your blueprint is a structured JSON document organized into 4 semantic layers, each containing 2–4 modules. Every module is individually editable, versioned, and distributable. Together they form a complete, machine-readable description of your identity.

Layer 1

Core Identity

The foundational layer. Captures who you are, what you do, and how you fundamentally prefer to work and communicate.

namerolecommunication_stylework_preferences

Layer 2

Voice & Tone

How you sound. Controls the register, density, and personality of AI responses — from terse and technical to warm and expansive.

formalityverbosityhumorpreferred_format

Layer 3

Code DNA

Your engineering fingerprint. Languages, stack preferences, code style rules, tool choices, and the guiding principles behind your technical decisions.

languagesframeworksstyletoolsphilosophy

Layer 4

Knowledge Base

Depth and breadth. The domains you know deeply, how you reason through problems, the mental models you default to, and your known blind spots.

domainsreasoning_stylemental_modelsbiases

Why this structure?

The 4-layer model maps directly to how AI models parse system prompts: persona (who), style (how), capabilities (what), and knowledge (why). Loading all 4 layers gives any LLM a complete picture without redundancy or conflicts between modules.

The interview

9 phases, one question at a time

The AI interview is not a form. It is a live conversation with an AI interviewer that adapts based on your previous answers. You will never see a checkbox grid. Each phase builds context for the next, so later phases can ask smarter, more targeted questions.

Identity & Role

Communication Style

Technical Background

Tool Preferences

Work Patterns

Problem-Solving Style

Domain Expertise

Collaboration Style

Values & Philosophy

Adaptive questioning

The AI adjusts depth and direction based on your answers. A developer who mentions Rust will get different follow-ups than one who mentions Excel.

Pause and resume

Your progress is saved automatically after every answer. Close the tab, come back tomorrow — it picks up exactly where you left off.

Encrypted in real time

Each answer is incorporated into your blueprint and encrypted before it is written to storage. The raw interview transcript is never persisted.

First API call

Fetch your blueprint

Once your interview is complete, your blueprint is available via the REST API. Generate an API key from your dashboard under Settings → API Keys and make your first request.

Request

bash

curl https://api.artificial-id.com/v1/blueprint \
  -H "Authorization: Bearer YOUR_API_KEY"

Response — 200 OK

json

{
  "id": "bp_01jd8x...",
  "version": 3,
  "created_at": "2026-03-01T10:22:00Z",
  "layers": {
    "core_identity": {
      "name": "Alex Chen",
      "role": "Senior Full-Stack Engineer",
      "communication_style": "direct, precise, context-aware",
      "work_preferences": ["async-first", "documentation-driven", "deep focus blocks"]
    },
    "voice_tone": {
      "formality": "semi-formal",
      "verbosity": "concise",
      "humor": "dry, occasional",
      "preferred_format": "structured with headers"
    },
    "code_dna": {
      "languages": ["TypeScript", "Python", "Rust"],
      "frameworks": ["Next.js", "FastAPI", "Axum"],
      "philosophy": "boring stack, exotic algorithms"
    },
    "knowledge_base": {
      "domains": ["distributed systems", "developer tooling", "AI/ML infrastructure"],
      "reasoning_style": "first-principles",
      "mental_models": ["second-order effects", "Chesterton's fence"]
    }
  }
}

Blueprint is decrypted server-side on request

When you make an authenticated API request, the server decrypts your blueprint using your key shard (combined with the shard from Google Drive) and returns the plaintext JSON. The plaintext is never cached — it is generated fresh for each authorized request.

Next steps

Keep exploring

Dive deeper into any part of the platform.

Ready to build?

Create your free account and complete the interview in under 20 minutes.

Create Free Account Back to Docs