The Relay, Not the Replacement: Human–AI Workflows
By the end you'll be able to
- Decompose a real task into steps and classify each by AI-fit: delegate, retain, or draft-then-review.
- Design handover points: what the human briefs downhill, what the human checks uphill.
- Evaluate a workflow by its weakest checkpoint, not its most impressive step.
- Predict
- Allocate
- Decide
- Field note
⚡ 30-second warm-up: from Lesson 17, what's the three-question test?
Is it checkable? Would the training diet plausibly cover it? What happens if it's wrong? Today that test becomes one half of a bigger design job: the "uphill baton" in a workflow you'll build yourself.
Two colleagues use the same AI for the same report. One saves three hours and ships their best work. The other ships a confident error to a client. Same tool, same model. Predict what differed.
Start with the wrong question
"Can AI do my job?" is almost unanswerable. Almost no real job is one task; it's a bundle of very different steps. The better question is which parts of that bundle each of you should run.
Commit to a prediction
The colleague who shipped a confident error most likely…
This lesson changes one thing: you'll stop asking whether AI can do "the job". Instead, you'll unbundle any task into legs, decide who runs each one, and engineer clean handovers between them.
Discover the mechanism
Almost every real task is really several tasks stitched together. Take a typical piece of client work. You understand the client, because that context is yours alone. AI gathers precedents, using retrieval, and then you spot-check its sources (Lesson 13). AI structures the options; you judge which one fits. You price it, because stakes and accountability are your call. AI writes a first pass from your brief (Lesson 11). You check every claim (Lesson 17). And you send it, because it's your name on it.
AI legs are bounded and checkable. Human legs carry context, judgement and consequences. Quality gets made or lost at the handovers: a rich brief going downhill, proportionate verification coming back uphill. Teams that thrive with AI aren't the ones with the best prompts. They're the ones with the best-designed baton passes, and a named human standing at the anchor leg.
Choosing which model "grade" runs an AI leg (fast and cheap, or the most capable) is a related skill; this lesson's Go deeper opens it up. Match the grade to the leg, not to the whole project. Lesson 21 completes the picture: what the final signature actually means.Work is a sequence of legs. The craft is choosing which runner runs which leg, and engineering clean baton passes between them. AI legs cover volume, drafts, transformation and first-pass search. Human legs cover framing, judgement calls, stakes decisions and accountability.
- Downhill baton is the brief: objective, context, constraints, examples (Lesson 11). Uphill baton is the check: verification proportionate to stakes (Lesson 17), or the tally test where people are affected (Lesson 18).
- Where it breaks: relay legs are fixed and run in sequence; real workflows loop and branch (agents, Lesson 16, run several legs internally, with the gates you already know to place). Runners tire too, but AI doesn't, and that's the trap: it tempts you to over-delegate judgement legs simply because the volume legs went so well. Fluent success at leg three is zero evidence of fitness for leg four.
Build a workflow yourself
Conceptual illustration: a deterministic consequences engine, not a live model. Pick a scenario. Allocate each step. Choose the check for anything AI touches. Then run the quarter and see what happens.
Where this bites at work
Luka redesigned his acquittal workflow around these legs. AI drafts sections from his structured brief, with the improvement loop capped at two rounds. Retrieval grounds every requirement in the funder's current guidelines. Luka personally verifies every reported figure against programme records. And he signs the acquittal himself; his name goes to the funder. Six hours becomes two, and quality goes up, not down.
A leg is worth automating when it earns four yeses: it recurs frequently, success is easy to describe explicitly, errors are cheap to catch and reverse, and the blast radius of a bad run stays bounded. Fewer than four yeses, and it stays manual; just revisit that call every quarter.
Retire the idea that using AI well just means writing better prompts. Prompting is one baton on one leg. The real skill is decomposition, allocation, gates and verification. It's a work-design skill, and anyone who runs a process can learn it. No code required.
Now make the call
Maya's firm is rolling out a client-facing AI assistant. It drafts excellent client responses. Someone suggests: "since it's doing so well, let's have it also decide loan pre-approval eligibility." What should Maya recommend?
Try asking the assistant whether it's ready, too. It's worth seeing that self-assessed readiness is still just generation.
Apply it to your work
Pick one recurring task. Write its steps down. Mark each one AI, human or hybrid. Define both batons for every hybrid step, and name the human anchor. One page like that is worth more than any prompt collection.
Prove it to yourself
Two quick questions, untimed and retryable. Get them right, along with completing the lesson and the decision, and you'll earn ★ Mastered. This is what stops "that made sense" evaporating by next week.
Explain it in your own words
Write two or three sentences for a colleague. Then compare them against the rubric, which checks your ideas, not your wording. Only you see your note.
✓ Lesson 20 stamped: Workflow Architect
You allocated a real workflow, ran it, and watched the same model produce wildly different outcomes depending on your design. One lesson remains: the one that makes all the others safe to use at full speed.
Evidence & review — how we know what this lesson claims
| Claim | Type | Basis | Review risk |
|---|---|---|---|
| Task decomposition and allocation, not prompt wording, is the primary driver of AI-assisted work quality | How it works | Human–AI complementarity research | Medium — reviewed quarterly |
| AI legs suit pattern-heavy, checkable work; human legs suit judgement, stakes and accountability | How it works | Series synthesis of Acts 3–4 | Low |
| Model "grade" should match the leg (cost × quality × risk), not the whole project | Practice-based guidance | Provider pricing and capability documentation | High — reviewed quarterly |
| Four-part automation-suitability test (frequency, explicit criteria, cheap-to-detect/reverse errors, bounded blast radius) | Practice-based guidance | Series synthesis | Low–Medium |
| The workflow relay builder | Teaching device | Deterministic consequences engine, labelled illustrative | Medium |
| The relay-race analogy | Useful mental model | Limits stated in the lesson | Low |
Concept review due: January 2027. Product-behaviour claims: quarterly.
Where you are in the machine
Human judgement now includes designing the whole relay, not just running one leg of it. Select any layer to see its role, its lessons and your live progress.
The machine, bottom to top: training data (2) → token pieces (3) → prediction engine (1, 4–9) → the desk (10–11) → senses (12) → live knowledge (13) → tools (14) → memory (15) → agent loops (16) → human judgement (17–21).
Next: the anchor leg
One lesson remains, and it's the one that makes all the others safe to use at full speed. It covers judgement, verification, and what responsibility means when a machine did the drafting.