Decomposing work
Using agents · topic using-agents/decomposition
Large tasks fail as single prompts. This topic teaches breaking work into pieces an agent can complete and you can check, iterating on the pieces as the results reveal what you actually needed, and recognizing the point where another round of delegation costs more than doing the step yourself.
Concepts
- Task decomposition
- Splitting a large piece of work into steps small enough that each has a clear input, a clear output and a way to check it. Good pieces can be handed to an agent one at a time, verified independently and redone without losing the others. Decomposition also surfaces the decisions that are yours to make before the agent starts filling gaps. glossary
- Iteration
- Working in rounds: delegate a piece, check it, adjust the brief with what you learned, delegate the next. The first result usually reveals a missing constraint or a wrong assumption about the goal. Treat each round as feedback on the brief rather than as the agent failing, and keep rounds short so mistakes stay cheap. glossary
- When to stop and do it yourself
- Recognizing the point where explaining, correcting and re-checking an agent costs more than doing the step directly. Signs include a third round on the same piece, a task that depends on knowledge only you hold, or a check that takes longer than the work. Taking the step back is a judgment call, not a failure of the tool or of you. glossary
Links
- Builds on: Delegating to an agent
- Leads to: Deciding and specifying
- Competencies drawing on it: Specifies work well enough for an agent to implement and for anyone to verify, Delegates a task to an agent and checks the result
Lessons
- Breaking a task into pieces an agent can finish (tutorial)
- Delegating in rounds (tutorial)
- When to stop delegating and do it yourself (explanation)
Your reference
Each lesson above adds its takeaways and its example here once you finish it. Your reference lists every lesson you have finished.
Breaking a task into pieces an agent can finish
Unlocks when you finish Breaking a task into pieces an agent can finish.
Takeaways
- A task given as one prompt comes back blended, with the gaps filled by the agent's own choices, and reading well is the only thing you can check.
- Split a large task into pieces that each have an input, an output, and a check. One source per piece, a fixed output, and a check you can tick.
- The split shows the decisions that are yours: which data goes in, what base a number uses, which of two options to show. Make them before the first brief, and write them into it as limits.
- A piece with an output but no check isn't ready to brief. Write the check first, and use it as the done-criteria.
- A good piece can be handed over on its own and checked on its own. When it fails, you redo that piece and the other two are unchanged.
Example
Write the one-page report on the bike pilot for the facilities manager.
[the three sources follow]
**Shared-bike pilot: report**
The pilot was a clear success. Usage grew steadily over the eight weeks,from 112 rides in the first week to around 190 a week by the end, and 80%of riders want to keep the scheme. The main complaint was that bikes weresometimes unavailable in the morning, which suggests adding capacity.
The price to continue is 1,800 per month on a twelve-month contract.Given the strong uptake and positive feedback, we recommend continuingand adding a second docking station near the east entrance.Delegating in rounds
Unlocks when you finish Delegating in rounds.
Takeaways
- Delegate one piece at a time, and check each result line by line against its criteria, pass or fail per line. A read of the whole result gives a feeling, and the lines give you the failing bullet.
- Read a miss as feedback on the brief. A miss often shows a limit that was looser than the check, such as a limit that allowed a count worked out from other counts. Change the description of the task with what the judging of the result showed.
- The fix goes into the brief as a limit, and into the brief of every later piece with the same gap. A reply in the chat stays in that conversation.
- Keep a log of two lines per round: what changed in the brief, and what the result showed. It is light enough to keep on a busy day.
- Short rounds keep mistakes cheap, because a miss is found in one small piece before anything else depends on it.
Example
Write the "Riders" section of a report from the survey note below. Format: the heading "Riders", then four bullets, one per finding, each count written as "n of 41 replies".
Use only the survey note as your source. Use the 41 replies as the base for every count, and don't turn counts into percentages. Don't add commentary or recommendations.
Done when: there are four bullets, each count is written as "n of 41 replies", every count is in the note, no percentage appears, and there is no commentary or recommendation.
[survey note follows]
When to stop delegating and do it yourself
Unlocks when you finish When to stop delegating and do it yourself.
Takeaways
- A round is a brief handed over, a result checked against the done-criteria, and a brief changed with what the check showed. Most pieces take one or two, and a first result with a gap is the normal shape of a round.
- A third round on one piece, a step that depends on knowledge only you hold, and a check that takes longer than the work are the signs that the next round costs more than the step done by hand.
- Taking the piece back is a judgment about cost. It says nothing about the tool, its quality, or your briefing, because none of those move where a fact is or how long a check takes.
- The piece you take back keeps the input, output and check the split gave it, and the other pieces stay delegated, so the split loses nothing.
- Stopping after spent rounds can feel like giving up. The spent rounds cost the same either way, and the log says whether the next step is yours, however stopping feels.
Example
This lesson has no runnable example or prompt block.
Sources
DLAI-11Agentic AI: M1 workflows and autonomy, M2 reflection, M4 evals and error analysis, M5 autonomous agents, DeepLearning.AI (course)Academy ai-fluency-framework-foundationsAI Fluency: Framework and foundations, Claude Academy (course)