Write your awesome label here.
Data and Models

Write prompts that get the answer first time, at the smallest footprint that does the job.

Eight modules on the Prompt Footprint and the Five Levers — specification, structure, reasoning depth, model selection and orchestration — for prompts you write yourself and for the agents that run without you.
8h 30m
of learning
8
modules
50 question
final assessment
85%
to pass
Credential
on completion

Effective, Efficient and Green AI Prompting, for Self & Agents

What
How
Tools
Who

How the course works

Five levers, applied in order

Specification, structure, reasoning depth, model selection and orchestration, each with the question to ask and the evidence to check.

For your own prompts and for agents

The same levers applied to what you write and to what runs autonomously, which behave differently at scale.

Grounded in SCI for AI

Footprint figures follow the ratified specification, with method and boundary stated.

What this course helps with

Understand what a prompt actually costs

A prompt is authored once and runs on every request. The first module accounts for that across cost, carbon, energy, water and embodied hardware.

Get the answer first time

Vague requests are retried, and every retry reruns the whole task. Specification is the lever that removes most of that work.

Keep agents inside a budget

Autonomous work multiplies calls and carried context. Budgets and stop conditions make the cost predictable.

Who this is for

  • ML engineers and architects designing AI features

  • Anyone whose prompts run at production volume

  • ML engineers and architects designing AI features

  • Anyone whose prompts run at production volume

Related courses:

For the training side of models rather than their use, see Green AI for ML Practitioners; for putting this guidance into coding tools, see Building a Green AI Agent.

What you will work with

The Prompt Footprint

Five dimensions of a single prompt, and what changes when it runs a million times.

The five levers

Each with its precondition, its trade-off, and the quality bar it must hold.

Procurement and disclosure

What to ask an AI supplier, and how to report your own figures without overstating them.

What you will be able to do

  • Account for a prompt’s footprint across cost, carbon, energy, water and embodied hardware
  • Apply the specification lever so a task succeeds first time instead of being retried
  • Control output structure and length, which is where much of the per-use cost sits
  • Match reasoning depth to task difficulty rather than leaving deliberation on by default
  • Route work to the smallest model that holds a written quality bar
  • Bound agentic work with budgets and stop conditions
  • Measure a prompt’s footprint and report it with its method and boundary
  • Ask the right questions of AI suppliers, and disclose your own figures accurately

Course Lessons

The full course

Every module ends with an application exercise and a worked solution.

01

Where ML’s Footprint Is Decided

A model’s footprint is created twice: once when it is trained, and again on every inference it serves. This module sets the decision metric and the amortisation rule that every later module is judged against.
module-01

02

Reuse, Adapt, or Train: The First Decision

The largest reduction available on the training side is not to train, yet the question is usually settled by habit before anyone computes it. This module compares six routes to a working model and what to ask of a base model before inheriting its cost.
module-02

03

Sizing the Model and the Data Budget

How large a model, trained on how much data, sets the energy term more directly than any later choice. This module covers compute-optimal sizing and how architecture changes training and serving energy independently of parameter count.
module-03

04

Data Efficiency: What the Model Is Trained On

What the training tokens actually are changes both how many are needed and how good the resulting model is. This is the layer where the footprint lever and the quality lever most often point in the same direction, and this module is careful about where that alignment breaks.
module-04

05

The Training Run: Execution, Stopping and Waste

This module examines the compute that produces nothing: epochs past convergence, crashes without a checkpoint, and experiments that repeat work already done. Waste at this layer is invisible in a cost report, because a failed run and a successful one look identical on the bill.
module-05

06

Search and Evaluation: The Two Multipliers

Earlier modules reduce the cost of one run; this one addresses how many runs happen, which is usually the larger number. It covers search budgets, comparing strategies by cost per answer, and evaluation as a cost that recurs for the life of the model.
module-06

07

Hardware-Aware Training

Compute demanded is one quantity; how efficiently it becomes energy is another. This module covers precision, batch size, accelerator fit and utilisation as the conversion factor sitting between the two.
module-07

08

When and Where to Train

This is the one lever that changes carbon without changing energy, by moving a run to a different hour or region. Training is often a strong candidate, and this module covers the constraints that decide it, along with water, preemptible capacity and embodied allocation.
module-08

09

From Weights to Deployable Artifact

This module deliberately spends training cost once in order to reduce serving cost on every inference that follows. It covers parameter-efficient adaptation, post-training and compression, and the repayment calculation that decides whether the trade is worth making.
module-09

10

Lifecycle, Retirement and Disclosure

Retraining cadence multiplies the training footprint, and retirement is the point at which every amortisation estimate becomes a measured fact. This closing module covers cadence, drift, registry and model card records, disclosure duties, and the capstone footprint report.
module-10

11

Final Assessment

A final assessment of 30 scenario-judgement questions - 80% to pass, 60 minutes, retake available.

Assessment & Credential

How you are assessed

  • A short knowledge check after every module, drawn at random from that module’s pool
  • A final assessment of 50 scenario-judgement questions from a pool of 200+
  • 85% to pass, 120 minutes, retake available
  • Questions test judgement in situations you will meet — not recall of figures that date

What you earn

GSF Digital Credential
Effective, Efficient and Green AI Prompting
  • Issued on passing the final assessment
  • Valid for 3 years, refreshed by a short update module
  • Shareable credential for your profile, CV and tender responses

Time to complete

Reading and exercises
8h 30m across 8 modules
Knowledge checks
About 5 minutes per module
Final assessment
120 minutes
Typical completion
1–2 weeks at a module per session, entirely self-paced

Join the Pilot

The GSF Academy is currently in pilot. If your organization would like to enrol learners and issue GSF certifications to your teams, tell us a little about what you need and we'll follow up to discuss access, pricing, and enrolment.

Request Access for Your Organization

First name
Last name
Work email
Tell us about your organisation
Thank you.

We will be in touch shortly