Closed beta for adults · free

You can code. Learn the AI part without starting over.

Courses start with what a variable is. AI-grafen starts with what you already know and builds the shortest path to what you need to deliver — with labs where tests decide whether the code holds up.

See what you already know — no account How we know what you know

Sound familiar?

You have followed a tutorial but don't know what you don't know.

Courses spend four weeks on things you already know.

ChatGPT answers what you ask — but not what you should have asked.

Pick a track

Build a RAG system you can trustEChunking, retrieval evaluation, citation and grounding — the whole chain from document to answer with traceable sources.Build an agent you can trustFArchitecture, planning, observability, safety and evaluation — and the coding agent that gets objective feedback from tests.Evals in practiceFLLM as judge, regression tests, statistical significance and how to value negative results — measurement that holds between releases.Fine-tune a model with LoRAFUnderstand why low-rank adapters work, choose target modules and rank, build a dataset, and evaluate the result with evals.AI in productionFServing, scaling, distributed systems, operations and calibrated uncertainty — plus a project where you deliver a service that holds up.Language models in practiceETokenisation, sampling, instruction models, structured output, hallucinations and Swedish — what you need in order to build with an LLM for real.

Labs along the way

You write the code in the browser or in a sandbox on the server. Hidden tests decide whether it holds up.

Lab: scaled dot-product attention with a causal maskDabout 60 min Lab: BM25 and a minimal RAG pipelineEabout 60 min Lab: an agent loop with tools, a budget and a traceEabout 60 min Lab: build an eval harnessEabout 60 min Lab: LoRA on a small network — train only the adaptersFabout 75 min Lab: tensors, autograd and your first training loopDabout 50 min

What you get

Diagnosis first

The questions follow the prerequisite chain back from the goal. You skip what you already know.

The shortest path

Only the knowledge you are missing, in the order it builds on itself. The path is recalculated as you learn.

Labs with hidden tests

You write the code. Tests you cannot see decide whether it holds up — not whether it looks right.

Proof of what you can do

Mastery requires several kinds of evidence. The certificate lists them and can be verified.

See where you stand — in ten minutes

No account. You see right away what you already know and where your path would start.

Where are you today?