Write your awesome label here.
Most of your code is now written with AI. You still have to know what makes it green.
The developer's job has shifted. You describe what you want, a tool produces it, and you decide whether to accept it, often faster than you can read it. That only works if you can tell efficient code from wasteful code. Fifteen modules on what makes software green, applied to the three things you now do: write it, generate it, review it.
18h
15
30 question
80%
Certification
Green Software Developer
How the course works
Four footprints
Runtime work, the demand your code creates, the generated code you accept, and your own development loop. Naming all four tells you which one a problem belongs to.
A demand model first
Five lines: trigger, frequency, growth, consumption, elasticity. Written before the work, it tells you whether the work is worth doing at all.
Patterns with the trade stated
Five fields: precondition, mechanism, trade-off, verification, cadence. Anything missing those is advice.
Why take this course
You are approving code you did not write
Generated code follows the average of what it learned from, and that average is full of a query per row, results with no limit, and retries with no ceiling. If you cannot name those patterns, you cannot ask for better or reject what comes back.
Green is not the same as fast
They often agree. Prefetching, sending two requests and taking the first reply, and keeping capacity warm all make things faster and all add work. The course covers the six cases where they disagree and how to decide.
Nothing tells you when you get it wrong
A slow service pages someone. A wasteful one just runs. There is no alert, so the check has to be built into what happens anyway: the default, the pipeline, the review.
Who this is for
- Developers writing production code, in any language
- Anyone generating a large share of their code with AI tools
- Reviewers approving more code than they could have written
- Tech leads deciding what the team's defaults should be
- No prior sustainability knowledge and no other GSF course needed.
Related courses:
- Green AI Prompting — the prompts and agents behind your AI features
- Building a Green AI Agent — putting this guidance inside your coding tools
- Green and Efficient Architect — the decisions above the code
- GreenOps Practitioner — running it day to day
- SCI for AI Practitioner — producing a figure you can defend
- Green Data Practitioner — the data, storage, queries and retrieval behind your code
What you will work with
A runnable lab, if you want it
A working service with ten deliberate defects. Measure it, fix them, measure again: 301 queries become 2, 140 polls become 37. Node only, no setup, no keys. Optional and not part of the credential.
Things you keep
A five-line demand model you write before you build a feature, a register of the changes you made and what each cost you, and a one-page capstone of evidence on one code path, ideally your own.
Measurement you can do today
You probably have no way to measure energy. The course uses what you do have (queries, bytes, calls and tokens) and is clear about what those numbers can and cannot prove.
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 & Certification
Time to complete