Agents make work arrive faster. The decisions still need an owner.
An experiment in
scaling judgment.
Completion pathRequires a decisionTravel is illustrative. Counts include every task.
0 tasks arrived, 0 completed, 0 waiting for human review, 0 in review.Run the workday to see how tasks move through your assumptions.
Compare human review for every task with a hypothetical 10% subset. Only the review fraction changes; your current scenario is pinned.
Illustrative policies. Neither is a SOC 2 requirement. The model does not assess risk or control effectiveness. Read the approval analysis
Average demand fits within available effort. Quality still needs its own evidence.
Nominal demand with evenly spaced arrivals.639 completed of 640 arrivals. 0 waiting and 1 in review.
48 reviewer-minutes of nominal steady demand per hour / 60 available. 64 of 640 tasks require review. Arrivals are spread through the day.Give an agent a job. Give its decisions an owner. Then look at what happens when the work accelerates.
In my note on building a personal agent team, I started with the seat contract and the approval boundary. Before adding another bot, define the job it owns, the evidence it brings back, and the point where someone has to decide. This lab follows that thread into a capacity question.
A boundary can be clear on paper and still become a queue. If more agent work arrives at that boundary, what changes on the other side? Do we have more time to inspect it? Is the decision easier to make? Or have we simply made the waiting less visible?
The visualization makes that tension inspectable. Mint tasks can finish while amber tasks accumulate. The total looks productive; the decision queue tells another part of the story. Both can be true at once.
I’m using a deliberately small model so its assumptions can be challenged. The purpose is to explore where the next unit of effort belongs: more agent output, more review capacity, or better preparation for the decision. It is not a safe staffing ratio, a forecast, or an argument that people must review everything forever.
The baseline produces 640 tasks across eight hours. Under its hypothetical review rules, 639 finish and one remains in review at 17:00. Double the agents at the same review capacity: 1,231 finish, 48 wait, and one remains in review.
Completion almost doubles. The review queue also grows. A total-output chart alone would hide the second result. These are outputs of this model, not measurements of a real team.
Keep 40 agents and the same review requirement. Either increase available review effort to 120 minutes per hour or assume each review takes three minutes instead of six. Both configurations finish 1,279 tasks, with none waiting and one still in review at the cutoff.
The outcomes match because the modeled service rate matches. The real costs may differ. An evidence packet, tests or a clearer scope could help a reviewer, but this experiment has not measured that saving or whether decision quality holds.
The burst scenario sends 40% of the day’s work into its first hour. It ends with the same completion and waiting counts as the baseline, but mean waiting among reviews that started is about 30 minutes instead of zero.
Look at time as well as totals. The mean excludes tasks still waiting; their count and oldest age are shown separately. Neither an empty queue nor a short wait establishes that the right decisions were made.
An approval policy creates review demand. The team still needs enough capacity to carry it out.
SOC 2 Type 2 is sometimes described as a ban on fully agentic merges. The reviewed evidence does not support that blanket claim. The change-management criterion, CC8.1, addresses authorization, testing, approval and controlled implementation. It does not prescribe a person clicking every merge button.
The controls an organization implements can be more specific. If its process requires an independent person’s approval, an agent bypassing that step breaks the stated control. A bot merging after the required approval is a different workflow. Whether an entirely automated approval path is appropriate depends on its risks, controls and evidence, assessed with the responsible control owner and auditor.
Type 2 examines how controls operated over a stated period. Generating code, reviewing it, authorizing a change, merging a branch and deploying production are separate actions to trace. A passing test suite or a second agent identity does not, by itself, establish effective approval or independence.
The workbench’s approval-policy presets change only the fraction requiring human review. Every task sets it to 100%; selected tasks sets it to an illustrative 10%. Your other inputs stay fixed and the previous scenario is pinned for comparison. Neither percentage is a SOC 2 requirement. The model assigns tasks mechanically; it does not classify risk, inspect approval evidence or measure missed defects.
Use the comparison to ask whether the workload fits a chosen policy. A shorter queue cannot tell you whether that policy is sound. Before changing a real approval boundary, examine the applicable controls and the quality of the decisions alongside waiting time.
Checked 5 October 2026. The full AICPA criteria and new AI TQA downloads require account access and were not read in this review. The interpretation draws on reproduced CC8.1 wording in an issued report and the attributed practitioner analysis. This lab does not assess SOC 2 compliance.
From the model to the work
Take the question into a real workflow before taking the numbers as advice.
Which output needs a person, what are they deciding, and who owns the consequence? Separate routine checks from consequential approval.
Record arrival, review start and completion times. Include work that is still waiting, and look at bursts rather than only the daily average.
Try a smaller batch, clearer evidence or more review time. Measure decision quality and rework alongside delay before changing the approval boundary.
These works frame the question. The lab adds an editable model; its starting inputs remain hypothetical.
Ongoing agents make work that continues to arrive a timely operational question.
An announcement does not establish review capacity or delivered business value.
Monitoring coverage, review latency and escalation describe different parts of oversight.
Internal reporting does not calibrate this model. Agent actions, transcripts and operational tasks are different units.
Publication date unverified.
Parallel agent work still needs verification, review effort and accountable ownership.
The page identifies an Anthropic affiliation; this is practitioner guidance, not independent validation of our simulation.
Checking source context before presenting a focused human decision suggests review preparation as another lever.
A direct account without a measured review-time saving.
Teams are exploring different review boundaries: risk, plans, tests, smaller changes and human review of agent feedback.
Reported approaches are context, not calibration. The paid remainder was not reviewed. This lab does not evaluate which boundary is safe.
The interview places review delay and human interactions alongside agent output. That gives this capacity question a concrete operational setting.
The reported timings are not independently verified, and are not used as this lab’s input values. The newsletter byline is Lenny; the workflow account is Zach Lloyd’s.
Selection and concurrent-agent time measurement complicate estimates of productivity.
Task flow alone cannot establish productivity, quality or business value; the update does not establish a precise current effect.
These are runs of this model. One final review remains in service at the cutoff; it is included in unfinished work.
| Scenario | Arrived | Completed | Waiting | In review | Mean wait* |
|---|---|---|---|---|---|
| Baseline | 640 | 639 | 0 | 1 | 0 min |
| Double the agents | 1,280 | 1,231 | 48 | 1 | 88.88 min |
| More review capacity | 1,280 | 1,279 | 0 | 1 | 0 min |
| Prepared review (assumed) | 1,280 | 1,279 | 0 | 1 | 0 min |
| Work arrives in bursts | 640 | 639 | 0 | 1 | 30.2 min |
*Among reviews that started. Still-waiting tasks are excluded from this mean and reported separately. Prepared review assumes three minutes of effort; it does not establish an observed saving.
