Product

Your data already knows.
CAAD helps you see it.

Convergent Anomaly and Aberration Detection. You have multiple data sources measuring the same thing. When they agree, trust the signal. When they don't — CAAD tells you where the gap is, why it exists, and how far it's spread.

How CAAD Works
Two observers. One truth. The gap is the insight.

Most anomaly detection asks: "is this reading unusual?" CAAD asks a different question: "do two independent ways of measuring the same thing agree?" That's a fundamentally harder question to fool.

01 — Connect
Bring your two sources

Any two independent measurements of the same system. Sensors and satellites. Field data and reported data. Model predictions and ground truth.

02 — Measure
We compute the gap

Not just "is there a gap?" — but where is it, how big, is it growing or shrinking, and is it caused by sensor error or something real?

03 — Decompose
Separate noise from signal

Your sensors have calibration drift. Your data has dropouts. CAAD separates these from genuine anomalies — so you don't chase ghosts.

04 — Act
Know where to look

First deviation point. Chain propagation. Coverage blind spots. You don't just know something is wrong — you know exactly what to do about it.

What CAAD Gives You
Not just detection. Diagnosis.
WHERE

First deviation mapping. The exact point in your system where the gap first appeared — not just "something is off" but "it started here."

WHY

Signal vs noise decomposition. Is this gap caused by a sensor glitch or a real event? CAAD quantifies both so you don't waste time on false positives.

HOW FAR

Chain propagation tracking. Once an anomaly starts, how far has it spread? Which neighbouring components are now affected?

WHAT'S MISSING

Coverage analysis. Your monitoring network has blind spots. CAAD measures their cost — so you know where to invest next.

Validated Domains
Same engine. Different systems. Proven results.

We validate CAAD against open-source, verifiable problems. Each study below uses real simulation tools and publicly available data. The results are interactive — explore them yourself.

Transport & Traffic

Road incident detection

Simulated 36-edge traffic network using SUMO. Blocked a road at minute 20. CAAD identified the exact segment, separated sensor noise from real congestion, and tracked the congestion wave across neighbouring roads.

Observer A: Ground truth vehicle counts
Observer B: Sensor network (70% coverage, 15% noise, 5% dropout)
Explore the study →
Weather & Climate

Cold front detection

72 hours of temperature monitoring across 24 Kenyan weather stations. When a cold front arrived, satellite estimates lagged behind ground readings. CAAD showed exactly which stations were in the anomaly zone.

Observer A: Ground weather stations
Observer B: Satellite-derived temperature estimates
Explore the study →
Seismic & Geothermal

Earthquake and aftershock localisation

120-minute monitoring window across the Rift Valley. M4.2 earthquake at minute 45, aftershock at minute 75. CAAD tracked wave propagation, identified monitoring blind spots, and separated instrument noise from real ground motion.

Observer A: Primary seismometers (precise, 60% coverage)
Observer B: Secondary accelerometers (noisy, 90% coverage)
Explore the study →

Your domain not listed? CAAD works anywhere you have two independent ways to observe the same system. Health, finance, agriculture, infrastructure — tell us what you're working with.

Work With Us
From pilot to production

We don't do demos with fake data. We start with your actual data sources, run a real analysis, and show you what your system is missing. If it works — and it will — we build from there.

Your data already has the answer.
Let's find it together.

Start with a 30-day pilot. Real data, real results, real insight into what your system is missing.

Book a pilot