Pilot · limited independent review. This is practice, not a certificate or a job guarantee. Before you start

DrillhallHands-on practice

Before you start

DeploymentLab · Pilot terms

Drillhall teaches you to do forward-deployed engineering work by doing it: reading a customer problem, making a decision, fixing real code, and explaining the result. You do not need to be an engineer already. You do need the basics below by the time you reach the hands-on labs.

What to bring

  • A laptop or desktop with a modern browser. The lessons and the starting-point check work in the browser alone.
  • A free GitHub account. The labs have you copy a starter repository into a public repository you own.
  • Git, and a terminal you can open and type in.
  • Python 3 for the Beginner Systems Lab (it runs python3 -m unittest).
  • Node.js 20 or newer for Project Atlas (its starter has no other dependencies).

What we assume you know

Nothing about FDE work. You should be able to read a short piece of code and follow a step-by-step instruction. Each lesson names anything it expects you to have seen earlier, and ends with links to authoritative reading for when you want more.

How the path runs

  1. FDE Core Primer.
  2. A starting-point check that assigns only the lessons you need.
  3. Foundation lessons, about 3½ hours if you take all eleven. Most learners are assigned fewer.
  4. The Beginner Systems Lab: clone, test, find a defect, push a verified repair.
  5. Project Atlas: the full customer journey, with assessed evidence.

Lab and Atlas time depends on your experience and is not estimated here. Work is saved, and you can pause and resume.

What this does and does not prove

Your results show practised, evidenced skills against a stated map. Passing is not a job guarantee, a certification, or a claim that you are ready to run production systems alone. Automated test checks are real. In this pilot, stakeholder replies and AI-judged rubric checks are scripted placeholders, labelled as such: they are not a real evaluation, and your work is not sent to an AI provider (see Privacy). If you think a result is wrong, email support from Contact with your run reference.

Never use real customer data or production credentials in any lab.