Skip to content

Advance Labs / Applied AI Systems

We make complex, real-world systems discoverable and operable by AI.

We build the software, data systems, and operating workflows that help organizations show up in AI-driven search and run complex operations with greater clarity, speed, and control.

Drone autonomy tests
275
Quantum emulator tests
228
AEO audit rules
54
Published guides
16
The toolkit

Find out what AI engines actually say about you.

Open source, Apache-2.0, and the same engine behind every audit we sell. Point it at a site and it scores what answer engines can retrieve, cite, and get right.

  • Project logo

    OPEN SOURCE / TOOLING

    AEO Toolkit

    An open suite of answer-engine-optimization tools and the rule engine behind them, built to measure whether a web presence is actually retrievable and citable by AI assistants rather than assuming it is.

    Crawls up to 50 pages and scores crawlability, metadata, schema, and answer-engine readiness out of 100. Free, no sign-up.

Open source

Getting your own name on the internet should not need a credit card.

runs-on.dev is a free web address we run. Sign in with GitHub, claim yourname.runs-on.dev, and you have a live page there in about a minute. Nothing to pay, nothing to renew. Every claimed name is written into a public file anyone can read, so you can always see who owns what, and point yours at your own site whenever you want.

runs-on.devOpen
The live registry. Claiming a name takes a GitHub account and nothing else. Open source, AGPL-3.0. Click the preview to use it right here.
What we do

Two practical ways AI changes how organizations compete.

01 / AI DISCOVERABILITY

Make your business the answer, not just a result.

Search is becoming conversational. We make sure AI engines can understand, retrieve, and cite what your business actually does.

  • Find where AI engines misread your business
  • Fix the content, entity, and technical signals behind it
  • Turn visibility into something you measure on a schedule

Built on our open-source AEO Toolkit. Open the toolkit

02 / AUTONOMOUS OPERATIONS

Make complex operations legible before you automate them.

We build the software layer around autonomous systems: operator interfaces, mission data, and decision support for teams working with drones and other real-world machines.

Current focus: partner-led R&D and software pilots, heading toward autonomous construction-site monitoring

  • Turn fragmented operational data into one system view
  • Design operator workflows around oversight and exceptions
  • Build foundations for mission planning, monitoring, and review

Built from open sim-first autonomy work, verified against PX4 SITL across 275 tests. See the simulation

About

A small lab, run by two brothers.

Advance Labs is an independent AI studio in London, Ontario. We're small on purpose, and we'd rather show you something running than talk about it. Everything on this page links to the thing behind it, so you can go and check.

Lucas keeps a running log of everything he builds. Explore Lucas's lab →

Proof of work

Built systems, not slide decks.

Project logoCLIENT / DELIVERED → JOINT VENTUREKuzyn Builds → BuildCode

Designed and built the marketing site for a London, Ontario custom-home builder, then set up its local AI-search visibility. The relationship grew into BuildCode: plain-English Ontario building-code lookup for the trades.

Introduced me to Claude, which has revolutionized my business.
Tom Kuzyn · Kuzyn Builds

2025 · London, Ontario

Read the case studyKuzyn on GitHub
Project logoCO-FOUNDED / SITE LIVE, ENGINE IN BUILDConverg3nce

An AI-visibility product for commerce brands: it reads a brand's web presence the way an AI shopping agent does, scores how agent-ready it is, and ranks the fixes.

2026 · Three co-founders

Visit the site
Method

Start with the system. Then build the leverage.

  1. 01Map

    Identify the information, workflows, incentives, and failure points that shape the real problem.

  2. 02Prove

    Build a narrow, measurable intervention before committing to a broad platform or transformation.

  3. 03Operate

    Ship the tools, measurement, and feedback loops needed for the system to improve in real conditions.

We do not sell AI theatre. We use automation only where it improves a decision, a workflow, or an outcome.

Research

The autonomy work, in the open.

The evidence under autonomous operations, published rather than described. It runs sim-first against PX4 SITL, the test count is the suite's own, and the source is publicly readable, so the claims can be checked rather than taken on trust.

  • Project logo

    PUBLIC / SIMULATION + CODE

    Urban Drone Autonomy

    A drone that plans and flies its own mission through a real city, in your browser. It decides where to go, and a separate safety layer can overrule any move it makes. The cities are real, imported from OpenStreetMap, and it works out where powerlines hang by spotting the poles that carry them. 275 tests against the PX4 flight stack. Open source, Apache-2.0.

    drone.advancelabs.devOpen
    Flying in the browser. Click the preview to take the controls: the drone picks its own route, and the safety layer can overrule it. Click the preview to use it right here.
Also in the lab

Built, working, and honest about where it stands.

Not everything gets attention at once. These are shipped and reachable, and the label on each one says what is actually true of it today.

Project logoPUBLIC / PAPER + CODEQuantum-Hybrid Research

Feasibility studies at the intersection of quantum and classical systems, including whether quantum hardware can accelerate LLM training. Every claim is tagged by the strength of its evidence, and the 228-test emulator behind the numbers is public.

Project logoOPERATED / PRODUCTIZINGHousehold

Chores, rent and a shared fund for a house of roommates. One source of truth behind a web app, an e-ink wall display and an MCP server. Run daily; being prepared as a self-serve product.

Project logoPRODUCT / PAUSEDDialed

AI phone agents for local-business call handling. Built and demonstrated, not currently maintained.

cartrix.live logoPRODUCT / PARKEDCartrix

A product experiment in checkout automation and purchase workflows at scale.

Live · The AI Visibility Wall

We ask ChatGPT, Gemini and Perplexity the questions your buyers actually type, then measure whether they name you, name a competitor, or say nothing at all. Every bubble is a real company sized by that score.

Check your own domain
Start here

Have a system AI should understand or help operate?

Tell us what is changing, where the workflow breaks down, and what a useful outcome would look like.

Already know what you need? See the AI-visibility audit or the fixed-scope build sprint.