AI-Assisted Engineering Live training from San Digital
Join the waiting list
Menu

Use AI coding tools and keep engineering standards high.

Live software engineering training from San Digital for engineers with two months to two decades of experience. Four 40-minute sessions use Python, TypeScript, C# and Rust exercises to help you test generated changes and make review decisions with confidence.

View the course Join the waiting list
Module 1 exercise verification

BUG REPORT

Imported transcripts contain duplicate segments when an upstream service retries an event.

ACCEPTANCE CHECK

Make ingest idempotent without broadening logs or unrelated code.

$ python -m unittest discover -s tests

2 tests passed

Review decisionThe visible tests pass. The change still needs independent checks before it can be merged.

You work on a supplied repository, run checks and decide what is safe to keep.

Four habits for day-to-day engineering work

The course is about the decisions around generated code: defining the task, checking the result, controlling the size of the change and recovering when an approach fails.

Define the work
Turn an unclear ticket into criteria, constraints and questions before code is written.
Check the result
Use tests, runtime behaviour and source documentation instead of trusting a generated explanation.
Control the change
Keep each delegated step small enough to inspect and reverse.
Recover cleanly
Stop repeating an approach when new evidence disproves it.

One engineering problem per session

Every module starts with a short briefing and moves into a prepared repository task.

  1. 1

    Verification before trust

    Treat generated work as untrusted until you have checked it independently.

  2. 2

    Problem definition

    Define ambiguous work before asking a coding tool to implement it.

  3. 3

    Bounded delegation

    Give a coding tool only the work you can still review properly.

  4. 4

    Recovery and change control

    Change course when the evidence shows that the current approach is wrong.

Read the full course outline

Lee Crossley

Meet Lee Crossley

Lee created the course, its repository exercises and the assessment material. He teaches every session live and leads the review after each task.

Lee has delivered hundreds of software projects across start-ups, consultancies and large organisations. He is a director at San Digital and an author of Production-Grade AI.

More about Lee and the teaching style

Join as an individual or bring a team

Individual course

Join the waiting list for dates, prices, priority booking and 10% off your first place.

Join the waiting list

Private course

Give your team a shared way to verify AI-assisted changes and own the software it delivers.

Training for your team