Code · architecture · study mode

Interactive Study Lab

You can put Python on your resume.
Can you name the skill you're building in AI?

Here you build one: LLM Training & Inference Engineering — the thing production AI engineers actually do. Every step comes four ways: watch it, read it, go deep with your tutor Rune, and beat the level.

We at FBA Labs are running the First Break Speedrun in FBA Lab — it's live now (Aug – Sep 2026).

Rune — an AI tutor that guides every step by voiceCode visualization & explanationEvent & gamification — a time-bound speedrun event
Community

Join the FBA Lab on Discord

Connect with learners working through the same labs. Get help, compare notes, and stay in the loop on cohort sessions and new content.

  • Ask questions while you work through labs
  • Share progress with cohort peers
  • Get updates on new tracks and releases
Join Discord
// tracks

Start with one lab track

Both tracks are self-contained. Start with speedrun if you want to understand how training improves a model. Start with inference if you want to understand how a model runs.

Phase 1 · Training

speedrun

Start with a baseline GPT-2. Watch six researchers' decisions — RoPE, Muon optimizer, logit capping, 32K context — each annotated in the training code and plotted on the loss curve. This is how models actually get better.

  • 6 iterations
  • 3.28 target loss
  • Python / nanoGPT
Start speedrun →Source: 6 runfiles on GitHub
Phase 2 · Inference

Qwen3 Pure C

The AI doesn't plan sentences. It picks one word, then the next — based on every word before it. Six stages trace how: loading the weight file, running attention, choosing the next token. ~1,200 lines of C. No black boxes.

  • 6 stages
  • ~40 blocks
  • C / GGUF
Start Qwen3 walkthrough →Source: run_viz.c on GitHub
fba-lab — bootready
FBA Lab study mode — pick a track to begin$ fba load-tracks: speedrun, qwen-csimulation ready — pick a lab track ✓
// preview

One block. Four synced views.

Pick a function on the left. The source code highlights in the center, the architecture diagram pulses on the right, the terminal logs what happened, and the explanation tells you why it matters.

Open the live lab →

Every step, four ways — all pointed at one skill

You choose how deep to go. The goal is the skill you're building, not the checkbox.

▶ Watch

A short video lesson walks you through what changed and why — before you open the code.

¶ Read

A written lesson for every step, at your own pace — skim it now, come back to it later.

✦ Go deeper with Rune

Optional deep-dives where your tutor Rune explains a concept in detail — trigger them only when you're curious.

◆ Play the speedrun

Acts, a cohort, and the loss curve as your scoreboard — every step is a level to clear.

Finish the cohort, earn your certificate

At the end of each event cohort, every participant can download a certificate naming the real AI skills they covered. Not a badge of attendance: a specific, defensible skill list.

🔒 Issued and downloadable by participants at the end of the cohort.

RepositoriesRepo — coming soon