AI-native software company · Sweden

We build AI-⁠powered apps and help teams put AI to work.

Addable Labs ships AI-powered products — largely built by AI agents with a person in charge — and helps companies and teams become fluent with AI, from assessment to adoption.

See what we have built Fluentor

  • Agent-built, human-directed
  • Open source where it can be
  • Gate-checked on every change
factory · build-basic · website
  1. requirements31 reqs, 31 criteriadone
  2. plandesign system, landingdone
  3. reviewplan reviewed, approveddone
  4. implement10 work items, 1 branchdone
  5. gatesten gates on the build10/10
  6. founderreviews and approveswaiting

Services

Three ways to work with us

01

AI-powered apps

AI as part of the product, not a bolt-on: the apps below were built largely by AI agents, with the founder setting the direction and approving what we publish. We build new apps only when they are clearly a job for an AI agent team.

What you get

  • A working product, not a slide deck
  • Source you own, with quality gates and CI from day one
  • An honest account of what AI can and cannot do for the product
See the apps

02

AI adoption

Find out where your people stand with AI, then move them: Fluentor self-assessments, a guide to the next level, and what we learn adopting AI ourselves, shared as we go.

What you get

  • A fluency baseline per person and per team
  • Experience and articles from our own AI adoption, shared as we learn
  • A guide to the next level for each person and the team
Visit Fluentor

03

AI experiments

We try things with AI agents — a game built from a single prompt, a website built end to end by an agent-run factory — and publish both what worked and what did not.

What you get

  • Honest evaluations, including the experiments that fell short
  • Notes on how the agent-run factory works, stage by stage
  • Open-source repositories you can read and run
Read the notes

Apps and products

What we have built

Four projects, from open-source apps to a product in preview. Status is stated as each repository states it.

AI adoption

in preview

Fluentor

Your AI journey, with Fluentor as mentor. In five minutes you see where you stand with AI in your work. Free for individuals. We let people in in rounds.

AI-powered apps

open source · MIT

Notesage

A rich-text (WYSIWYG) Markdown editor that collaborates with AI. A native desktop app that transcribes speech on the device.

AI experiments

experiment

Ashlands

An action-RPG that AI agents built from a single prompt, steered by a person along the way, with an honest evaluation. An experiment, not a product.

AI-powered apps

open source · MIT

Gaimer

Describe a game in plain text and get a playable HTML5 game in seconds. A desktop app on Tauri 2, Vue 3 and Quasar with OpenAI or Claude.

How we work

AI-native, agents do the building, a person is accountable

How this site was built by agents See the open-source portfolio

AI-native
AI is part of what we build and part of how we build. The products on this page were made largely with AI agents.
Agents do the work
The founder sets the direction and reads every text before it is published; agents plan, implement and review the work in between.
Proof, not claims

This very site was designed, built and is maintained by an agent-run software factory. The founder reads every text before it is published.

A Swedish company
Founded by Péter Blénessy. Public repositories and honest evaluations — including the experiments that fell short.

Latest writing

Notes from the work

  1. From self-assessment to AI journey: how Fluentor came about

    In November 2026 nivå became Fluentor. The product had moved on to what comes after the assessment, while its name, logo and page still talked about your level. This had to be changed.

  2. Set the limits in advance

    A line in the instruction file asks an AI agent to behave, while a limit is checked outside it. This part goes through the limits the tools offer, a first set to try on a practice project and what to do when a limit lets something through. The seventh part of the series "From AI chat to agentic work".

  3. Where the AI agent runs and who starts it

    An agent that runs with nobody beside it cannot ask anyone, so a good first job to run on its own is one that can only read. This part follows the same agent to five places, from the editor to a schedule and the vendor's cloud. It says what goes along to a run in the cloud and ends with a first job that runs by itself every morning. The sixth part of the series "From AI chat to agentic work".

All articles

Contact

Have an idea?

Tell us about the app you have in mind, where your team stands with AI, or a project you would hand to an AI agent team. We reply within a couple of days.

Reach out hello@addablelabs.se