Write your awesome label here.
Run green software and AI as a gated, measured practice, from requirements to retirement.
Carbon, energy, water and waste managed together at every stage of the lifecycle and every layer of the stack: a green gate at each software and AI stage, a footprint budget, reduction work verified one change at a time, and one register read at a quarterly review.
10h 30m
13
30 question
80%
Certification
GreenOps Practitioner
How the course works
Gated rather than aspirational
Practices, gates and owners placed in the reviews and pipelines that already exist, modeled on the quality and security gates teams already run.
Pairs with FinOps
A shared allocation core and tagging, with cost and footprint read side by side as two ledgers, never merged, and reviews coordinated rather than duplicated.
Verification in every practice
The measurement that confirms a change is part of the practice, not an afterthought.
What this course helps with
A checkpoint at every stage
A green gate at the end of each stage states what is measured, the pass criterion and who signs it off, so footprint is designed in rather than found afterwards.
Improvements you can evidence
Each attributable change is measured before and after on a held basis, so its effect is confirmed or refuted.
Signals that reach someone who can act
An owner for each gate and budget, so an exceeded budget leads to a recorded decision rather than a silent exceedance.
Who this is for
- Engineers and architects embedding green practice across the lifecycle
- Requirements, design and code reviewers, and product owners
- Platform, operations, SRE and DevOps teams
- Sustainability practitioners, and practitioners adopting GreenOps alongside FinOps
Related courses:
SCI for AI Practitioner is the hands-on companion: it produces the measurements this course teaches you to commission and read.
What you will work with
The operating cadence
What is reviewed monthly (or every sprint), quarterly, and when an external obligation requires it, and who owns the response when a figure moves.
The register
One shared record of budgets, gate results, measurement statements and decisions, written so someone outside the team can read it.
Reduction in production
The operations loop on the running feature: demand review, right-sizing, scaling, inference and carbon-awareness levers, each change quality-checked and made one at a time.
What you will be able to do
Course Lessons
The full course
01
The Prompt Footprint
Introduces the five-dimensional view: cost, carbon, energy, water, embodied hardware. Establishes SCI for AI as the standards anchor. Positions the regulatory landscape (CSRD, EU AI Act, procurement disclosure) as the why now for Footprint literacy.
Before You Begin
Modules
02
The Specification Lever
How prompt intent quality drives Footprint. The GSF Intent Framework introduced. Trade-offs: over-specification wastes tokens, under-specification wastes retries. Worked examples across capability tiers.
Modules
03
The Structure Lever
Prompt organization: system messages, few-shot examples, output constraints, formatting instructions. Where structure earns its tokens and where it doesn't.
Modules
04
The Reasoning Depth Lever
When to invoke chain-of-thought, tree-of-thought, extended reasoning modes; when not to. The GSF Reasoning Budget: task-class taxonomy for reasoning depth decisions.
Modules
05
The Model Selection Lever
The most consequential single decision in inference efficiency. Capability tiers and routing decisions. The GSF Model Fit Test protocol. Same task across tiers, measured.
Halfway Check-in
Modules
06
The Orchestration Lever
Where agents leak resource: tool selection, chain depth, redundant retrieval, missing caching. Token budgets as a first-class design constraint. Case study from the supply-chain agent paper (70–90% workflow-level savings).
Modules
07
Measuring the Prompt Footprint
Hands-on demonstrations with EcoLogits, CodeCarbon, AI Energy Score. Producing numbers a CSRD auditor can read. The measurement chain from token count to gCO₂eq.
Modules
08
Procurement & Disclosure
What to ask your LLM vendor. What CSRD and the EU AI Act will require. The GSF Procurement question bank organized by Footprint dimension.
Modules
09
Course Completion
50 questions, 120 minutes. Pass Marks: 85%
Final Assessment
Share Your Feedback
Assessment & Credential
Time to complete