Life-phase and Milestone Adaptation

By learning from user events, documents, and temporal cues, Fastino helps your AI systems stay relevant and supportive as a user’s life evolves.

Overview

The world model maintained by the Fastino Personalization API continuously absorbs context from events, documents, and conversations.
Over time, this data reveals a user’s life-phase (e.g., student, early-career, founder, parent) and milestones (e.g., new job, move, launch, retirement).

Agents can then adapt tone, scheduling logic, and recommendations automatically — without manual rule updates or retraining.

Key Dimensions

Example: Inferring a Life-phase

Fastino can infer phase transitions automatically by analyzing recent ingested documents or events.

Input

POST /ingest
{
  "user_id": "usr_42af7c",
  "source": "notion",
  "documents": [
    {
      "doc_id": "doc_99",
      "kind": "note",
      "title": "New role announcement",
      "content": "Excited to join Fastino as Head of Engineering! Moving to San Francisco next week."
    }
  ]
}

Result

When you later query for user context:

POST /query
{
  "user_id": "usr_42af7c",
  "question": "What recent life transitions has Ash experienced?"
}

Response

The agent can now adapt scheduling, tone, and recommendations accordingly.

Example: Adapting Tone and Recommendations

Agents can use life-phase insights to personalize tone and actions dynamically:

Example: Milestone-Triggered Summaries

Fastino generates phase-aware summaries using the /summary endpoint with a purpose value like life-phase or transitions.

Response

This summary can be passed to downstream LLMs as part of system context, ensuring reasoning is phase-aligned.

Capturing Explicit Milestones

You can also log explicit transitions through the ingestion API.

POST /ingest
{
  "user_id": "usr_42af7c",
  "source": "user_profile",
  "events": [
    {
      "event_id": "evt_milestone_1",
      "type": "life_event",
      "timestamp": "2025-10-27T09:00:00Z",
      "metadata": { "category": "career" },
      "content": "User promoted to VP of Product."
    }
  ]
}

Agents that query this user will automatically incorporate that context into reasoning and decision-making.

Combining with Other Use-cases

Life-phase adaptation often interacts with other personalization capabilities:

Example: Phase-Aware Scheduling

1. Fastino detects a milestone: “Started new job.”

2. The scheduling agent queries:

3. Fastino replies:

4. The assistant automatically adjusts its scheduling heuristics.

Integrating into Agent Architectures

Example Implementation (Python)

import requests

BASE_URL = "https://api.fastino.ai"
HEADERS = {"Authorization": "x-api-key sk_live_456", "Content-Type": "application/json"}

def update_milestone(user_id, title, content):
    payload = {
        "user_id": user_id,
        "source": "career_agent",
        "events": [
            {
                "event_id": f"evt_{title.replace(' ', '_')}",
                "type": "life_event",
                "timestamp": "2025-10-27T10:00:00Z",
                "content": content
            }
        ]
    }
    return requests.post(f"{BASE_URL}/ingest", json=payload, headers=HEADERS).json()

def get_phase_summary(user_id):
    r = requests.get(f"{BASE_URL}/summary?user_id={user_id}&purpose=life-phase", headers=HEADERS)
    return r.json()["summary"]

# Example usage
update_milestone("usr_42af7c", "Promotion", "User promoted to VP of Product.")
print(get_phase_summary("usr_42af7c"))

Use Cases

Best Practices

• Include timestamps for all milestone events — recency improves relevance.

• Query purpose=life-phase summaries before major decisions or scheduling.

• Combine explicit user input (“I just moved”) with inferred signals from text or tools.

• Use /query to verify uncertain life-phase inferences.

• Regularly prune outdated milestones using /delete?scope=source.

• Respect privacy: never infer sensitive personal data without user consent.

Example: Deterministic Life-phase Summary

Response

This deterministic summary can be embedded into LLM prompts or used to condition agent reasoning dynamically.

Summary

The Life-phase and Milestone Adaptation use-case allows your agents to remain empathetic, context-aware, and evolution-ready.
By feeding milestone data and inferred transitions into Fastino’s world model, your assistants gain the ability to recognize change, adapt tone, and realign goals — staying continuously relevant as users evolve.

Next, continue to Personalization Use Cases → Routine Prediction to learn how Fastino anticipates user activity patterns and energy cycles.