System Active — Backfill in Progress

AURORA

Autonomous Understanding & Retrieval of Observed Activity

A complete digital archaeology of a human life. Sixteen years of personal data — ingested, cross-referenced, and analyzed by seven AI engines to surface patterns invisible to human perception.

11.9M
Total Records
22+
Data Sources
3,507
Behavioral Signatures
~6,000
Days of Fingerprints

The Numbers

Every data point is a thread. AURORA weaves them into a tapestry of behavioral understanding that spans 16 years and counting.

6.9M
Scene Classifications
3,441 scene types identified across 117K photos via computer vision.
4.0M
Health Records
Steps, heart rate, sleep, blood pressure, VO2 max, walking biomechanics.
512K
Google Activity Events
273K searches, 141K maps, 50K YouTube, 8K flights.
36,568
Emails Archived
22 years of email across 3,185 unique sender domains.
246
Flights Reconstructed
94 airports, spanning 2010 to 2026.
2.3M
Words Transcribed
626 recordings, 238 unique speakers identified.

The Data Universe

Twenty-two data sources, six categories. Each source is ingested, normalized, timestamped, and cross-indexed against every other.

Photo Scenes
6,949,283
Apple Health
4,013,435
Google Activity
512,583
Photo Faces
115,595
Google Voice
68,054
Knowledge Graph
40,339
Facebook
38,498
Instagram
36,828
Gmail Archive
36,568
Google Timeline
34,174
Google Chat
29,300
Apple Personalization
17,869
knowledgeC
13,092
Gmail Insights
7,074
Safari History
4,700
Health Summaries
3,006
Looki Moments
1,624
Plaud Recordings
626
Strava Activities
363
Travel Places
340
Health Workouts
271
Flights
246
Health & Fitness
4,017,075 records
Apple Health Records
4,013,435
Health Summaries
3,006
Strava Activities
363
Health Workouts
271
Photos & Media
7,065,504 records
Photo Scene Classifications
6,949,2833,441 types
Photo Face Detections
115,595161 people
Plaud Recordings
6262.3M words
Communications
140,996 records
Google Voice
68,054calls/SMS/VM
Gmail Archive
36,5683,185 domains
Google Chat
29,300
Gmail Insights
7,074purchases/trips
Social Networks
75,326 records
Facebook
38,498posts/msgs/reactions
Instagram
36,828likes/msgs/follows
Location & Travel
34,760 records
Google Timeline
34,174GPS 2009-2025
Travel Places
340
Flights
24694 airports
Digital Activity & Intelligence
598,476 records
Google Activity
512,583search/maps/YT
Apple Intelligence Graph
40,339knowledge triples
Apple Personalization
17,869interest scores
knowledgeC
13,092app usage
Safari History
4,700
Looki Moments
1,624wearable AI
Ingestion Pipeline
Ingest
22+
Raw data sources
2009 – present
Normalize
Classify
11.9M
Records processed
by 7 AI engines
Correlate
Synthesize
3,507
Behavioral signatures
across 13 dimensions

16 Years of Data

From the earliest Google search in 2009 to this morning's heart rate reading. Approximately 6,000 days of continuous behavioral fingerprints.

'09
'10
'11
'12
'13
'14
'15
'16
'17
'18
'19
'20
'21
'22
'23
'24
'25
'26

The AI Orchestra

Seven specialized AI engines, each assigned a role. No single model can do it all — the orchestra plays in concert, each instrument chosen for what it does best.

Reasoning & Narrative
Claude Sonnet
Anthropic
Deep reasoning, narrative generation, behavioral analysis, cross-domain insight synthesis.
Fast Classification
Claude Haiku
Anthropic
Lightweight classification, quick extraction, high-volume triage of incoming data.
Gating & Classification
Gemini Flash
Google
Significance gating, change classification, pattern detection at near-zero cost.
Photo Vision
Amazon Nova Pro
AWS Bedrock
Scene understanding, object detection, face recognition across 117K photos.
Sentiment Analysis
AWS Comprehend
Amazon Web Services
Emotional tone extraction across email, messages, and social interactions.
Predictive Models
Amazon Forecast
Amazon Web Services
Time-series forecasting for health trends, behavioral cycles, and seasonal patterns.
Transcription
OpenAI Whisper
OpenAI
2.3 million words transcribed from 626 recordings with speaker diarization.

13 Dimensions

Every day is scored across thirteen behavioral dimensions, computed nightly. Together, they form a high-dimensional fingerprint of lived experience.

01
Health
4.0M records
Apple Health, Strava
02
Photos
7.1M classifications
Scenes, Faces, Albums
03
Communication
141K messages
Gmail, Voice, Chat
04
Search Behavior
273K queries
Google Activity
05
Location
34K GPS points
Timeline, Flights, Places
06
Social Media
75K interactions
Facebook, Instagram
07
Schedule
calendars + email
Calendar, Gmail Insights
08
Digital Activity
18K events
knowledgeC, Safari
09
Voice
2.3M words
Plaud, Google Voice
10
Fitness
634 activities
Strava, Workouts, Rings
11
Motion
activity + steps
CoreMotion, Health
12
Music
listening history
Apple Music, Spotify
13
ARIA Meta
self-referential
Conversations, Journal

Discoveries

What happens when you cross-reference every dimension of a life against every other? Patterns emerge that no single data source could reveal.

01
Health × Schedule Correlation
Meeting Load Predicts Resting Heart Rate and Sleep Quality
AURORA detects that resting heart rate increases by 4 BPM on days with four or more meetings, and the effect persists into the next night's sleep quality. The body keeps score of the calendar.
Apple Health Calendar Health Summaries
02
Travel × Communication × Health
Complete Trip Reconstruction from Fragmented Data
Cross-referencing 246 flights with 34,174 GPS points and 36,568 emails, AURORA reconstructs every trip since 2010 — destinations, durations, who was communicated with during each trip, and how health metrics shifted across time zones.
Flights Google Timeline Gmail Archive Apple Health
03
Search × Health × Fitness
Physical Activity Drives Digital Food Exploration
By analyzing 273,000 Google searches alongside health data, AURORA discovers food and restaurant search frequency is 3x higher on high-step-count days. The body's physical activity triggers a predictable pattern of culinary curiosity.
Google Activity Apple Health Strava
04
Voice × Schedule × Communication
Leadership Transition Detected in Meeting Dynamics
626 Plaud recordings with 238 identified speakers reveal that average meeting duration increased 40% in Q1 while speaking ratio decreased — AURORA surfaces a leadership transition pattern: listening more, directing less.
Plaud Recordings Calendar Gmail
05
Communication × Social × Health
Emotional Sentiment Precedes Physiological Change by 48 Hours
The emotional arc across seven communication channels — messages, emails, social media — shows sentiment shifts that precede measurable health changes by 48 hours. The mind moves before the body follows.
Gmail Google Chat Facebook Instagram Apple Health
This is not a dashboard.
It is a digital archaeology of a human life.

Eleven million records. Seven AI engines. Thirteen dimensions. Sixteen years. One person. Every data point a fragment. Together, they form the most complete computational portrait of an individual life ever assembled outside a research lab.

Project AURORA — Built on the ARIA Platform — 2024–2026