OBD-II · AI · Cloud

Car Diagnostics Platform

On-board OBD-II diagnostics with cloud analytics, AI assistant and fleet support.

Check car compatibility
Possible situations

What our users actually find

Sample scenarios based on common cases — real customer quotes will appear once we collect publication consent.

What you'll see

The whole story on one screen

DTC + plain-English explanation

Why the light came on, how serious it is, and what to do next — no jargon.

Live telemetry

RPM, coolant temp, MAF, throttle — in real time, like a factory service manual.

AI assistant

Describe the symptom — the model cross-checks ECU data and suggests the likely cause.

Do the math

How much you'll save per year

Shop diagnostic visits
$60 – $120 × 2–3 times/year
≈ $180 – $360
App + ELM327 adapter (one-off)
≈ $10 (adapter) + subscription from $3/month
≈ $46 / year
Difference
≈ $130 – $310 per year

Estimate for a single-car driver who deals with 2–3 unexplained warnings a year. Your savings depend on how often things go wrong.

Compatibility

Will your car work?

OBD-II is mandatory on passenger cars (USA since 1996, EU since 2001) and on light commercial vehicles (since 2005).

Example session

What AI diagnostics looks like on a real code

2018 Ford Focus 1.5 EcoBoost Code P0301 Cylinder 1 misfire detected
  1. 01

    What the scan sees

    DTC P0301 confirmed, live data shows a +12% Long-Term Fuel Trim spike on cylinder 1, brief misfires under load.

  2. 02

    What the AI says

    Most likely cause — cylinder 1 ignition coil (44%), then spark plug (28%). Injector — 15%, compression — 10%. Recommendation: swap the coil first; the AI predicts P0301 will clear within 3 cold starts.

  3. 03

    What to do next

    Coil ~$45, 20-min DIY or ~$30 labour at a shop. Vs. a $150 scan-and-guess shop diagnostic — getting the right part first saves time and money.

Illustrative example — real findings depend on your car.

What powers the AI

How our AI diagnostics work

Specialized knowledge base

Approximately 145,000 DTC codes with expert descriptions, stored server-side — never bundled into the mobile app. The AI is grounded in this data rather than answering from general training knowledge.

RAG: retrieval-augmented generation

A dedicated retrieval service backed by a vector database with text embeddings. Relevant technical material is retrieved and injected into the model context before it answers — the model does not answer from its general training knowledge alone.

The model

A large commercial language model accessed through a model-routing abstraction layer, so the provider can be swapped without altering the diagnostic pipeline.

Thin client by design

The mobile app is an Apache Cordova shell with a React UI layer. The diagnostic session is server-owned: the phone acts as a transparent proxy, not the decision-maker. Logic lives once, server-side — auditable and upgradable without an app-store release cycle.

By the numbers
  • 1900+ DTC codes explained
  • 1996+ Years of OBD-II vehicles
  • 8 AI agents for diagnostics
Our limits

What we won't do

No ECU flashing without consent

We read parameters and DTCs (Mode 01, 03, 09) — passive queries only. Writing to the ECU only happens with explicit consent and a risk disclosure.

We recommend a shop when needed

When a code points to pressurized fuel systems, brakes or airbags, the AI plainly says "go to a shop" — instead of pushing DIY to keep you subscribed.

No fake urgency, no upsell scripts

No "3 seats left" or "special price today". The tier price is the same tomorrow as it is now.

Most-asked questions

Quick answers

  • Do I need to buy anything besides the app?

    Yes, an OBD-II adapter on the ELM327 chip (Bluetooth Classic SPP) — starts around $10. We do not sell adapters, but we maintain a compatibility list in the docs.

  • Which vehicles are supported?

    All cars and light commercial vehicles with an OBD-II port — mandatory since 1996 in the US, 2001 in the EU. Heavy trucks on J1939 are on the roadmap.

  • Is connecting to the ECU safe?

    Reading data (Mode 01, 03, 09) is fully safe — passive queries only. Clearing DTCs (Mode 04) is safe but requires explicit confirmation. We do not perform write operations without certification.

Download