Write your awesome label here.
Design agentic systems that finish the task on fewer steps, fewer tokens and less energy.
One completed task as the unit, eleven design levers, grounded in SCI for AI. From whether the agent is needed at all, through decomposition, tool surface, context, retrieval, termination, routing, failure and placement, to proving the change — plus an optional lab where you watch the number move, and a capstone on one of your own features.
10-12h
13
30 questions
80%
Certification
Designing Green Agentic Systems
How the course works
Framework-neutral
The levers apply whatever you build on. One worked implementation illustrates them; nothing depends on a particular agent framework.
One system throughout
The Northbridge support-triage agent, at 4,200 tickets a day, carries every worked example, every worked answer and the lab.
An optional lab where the numbers are live
Ten exercises against a running agent; the nine with worked fixes take the baseline from 2.925 to 0.362 gCO₂e per completed task. Every figure printed in the module is an output of the code that ships.
What this course helps with
The loop you did not need to build
Much of what an agent spends is a deterministic process wearing a loop — steps that were known in advance, rediscovered on every request. Module 2 asks for the number instead of the opinion.
Consumption that accumulates per step
Context that grows every turn, tool definitions re-read, retrieval that returns more than it needs, retries that repeat a step with its whole accumulated context. Energy, and cost, rise together.
A figure you can show
Energy and carbon per completed task, with the boundary, method and date recorded — so the reduction can be checked rather than asserted.
Who this is for
- Architects designing agentic systems and developers building them
- Teams putting an agent into production and answering for its cost and footprint
- Platform engineers setting defaults for agents across several teams
- Assumes working familiarity with LLM applications; no measurement background required
Related courses:
For how to obtain the numbers this course asks you to move, see SCI for AI Practitioner: The Measurement Lab. For fleet-level operations and placement policy, see GreenOps Practitioner. For the water dimension of a placement change, see Software Water Essentials.
What you will work with
The agent decision test
Sample real runs, count distinct sequences, cost the deterministic version, and write the benefit sentence for whatever remains.
The accumulation curve
How context, tools and retrieval compound across a run, and where compaction, projection and scoping cut in.
The lab harness
Prefill and decode reported separately, along with steps, tool calls, retrieved chunks, retries and cost per completed task. A Software Carbon Intensity (SCI) figure is calculated from stated coefficients.
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