QuixLab early access is open for engineers who want superpowersGET EARLY ACCESS

QuixLab

An agentic, collaborative notebook for analysing and reporting on large engineering datasets

01 UPLOAD

Upload your files.

Create an upload cell and drop your MF4 logs into it. You can look at the data before anyone builds an ingestion pipeline.

Watch the tutorial · 1:32

02 DECODE

Decode every signal.

Create an MF4 decoder from the upload cell. It reads the DBC files embedded in the log, or a DBC file you connect, and writes one row per signal sample.

One 2 MB log, 322,358 rows.

Watch the tutorial · 1:32

03 FILTER AND VISUALISE

Filter out the noise.

Ask an AI cell, in plain English, to filter for the data you need.

Then split the values by signal and plot them.

Watch the tutorial · 0:54

04 INVESTIGATE

Analyse with AI assistance.

Describe issues to an AI agent, which will write and run its own analysis.

The near-zero values are a decoding problem rather than a sensor fault. Two CAN buses carry the same speed frame, and one of them is decoded wrongly.

Keep the speed from Powertrain HS-CAN 1 only, and the trace is clean.

Watch the tutorial · 3:10

05 SAVE

Store your results.

Process multiple files using the same pipeline. Save the results to a Quix Lakehouse table, partitioned by platform, device, route and signal.

Each file is saved once, as its own batch.

Watch the tutorial · 2:09

06 AUTOMATE

Process new files automatically.

Switch the decoder from manual to automatic. The same cells now run as a pipeline, and every new log is decoded, filtered and saved without anyone opening it.

Each run is recorded with its inputs and outputs, so you can check what happened.

Watch the tutorial · 1:20

07 QUERY

Query every row.

Ask the lakehouse for all the data from one vehicle. QuixLab joins the rows from every file into one dataset, without opening a single file.

Watch the tutorial · 1:24

QuixLab is in early access

QuixLab pricing

Monthly, or annual at 17% off. Prices include VAT.

Starter

Build and run your first analysis workflows on your own test files.

£25 /mo · 1 user billed annually

2,500

cell executions a month

START NOW

Hosted by Quix

This plan includes:

  • 1 workspace
  • 1 worker
  • 1 GB load per execution
  • 50 GB of test data storage
  • 20m AI credits per month
  • 7 days of run history
  • Quix Analytics Agent
  • Community forum support

Business

For engineering teams of up to 100 people who need QuixLab in their own cloud, with controls.

£678 /mo billed annually

cell executions a month

START NOW

Self-hosted

Everything in Pro, plus:

  • 50 workspaces
  • Up to 20 parallel workers
  • 10 GB loaded per execution
  • Unlimited test data storage in your own bucket (S3, Azure, GCS, MinIO).
  • Pooled AI credits across the team
  • 365 days of run history
  • SSO (SAML, OIDC)
  • AI cost controls per user group
  • Your own Git repository (GitHub, GitLab, Bitbucket, Azure DevOps)
  • QuixDrive for seamless data ingestion from your network drives
  • Guest access for suppliers
  • Dedicated support

Enterprise

For organisations that answer to certification, security and procurement.

Contact Sales

Custom

cell executions

CONTACT SALES

Hosted by Quix or self-hosted

Everything in Business, plus:

  • Unlimited workspaces
  • 300+ parallel workers
  • Bring your own AI key
  • Development and production environments
  • Multi-cluster hosting
  • Air-gapped hosting
  • Custom networking
  • Audit log
  • RBAC with fine grained permissions
  • Dedicated support with SLA, onboarding and engineering support