<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Veronica Stewart</title>
    <description>The latest articles on DEV Community by Veronica Stewart (@veronicstewart).</description>
    <link>https://dev.to/veronicstewart</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4218065%2Fea07af82-0631-46a5-9a52-f9c3c61cb49d.png</url>
      <title>DEV Community: Veronica Stewart</title>
      <link>https://dev.to/veronicstewart</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/veronicstewart"/>
    <language>en</language>
    <item>
      <title>The Dive: the verification step your agent batch is missing</title>
      <dc:creator>Veronica Stewart</dc:creator>
      <pubDate>Fri, 30 Oct 2026 23:50:16 +0000</pubDate>
      <link>https://dev.to/veronicstewart/the-dive-the-verification-step-your-agent-batch-is-missing-4e8i</link>
      <guid>https://dev.to/veronicstewart/the-dive-the-verification-step-your-agent-batch-is-missing-4e8i</guid>
      <description>&lt;p&gt;Your agent pipeline can be composed cleanly, presented beautifully, executed exactly, and reconciled to the digit — and still be wrong, item after item.&lt;/p&gt;

&lt;p&gt;I learned this running a real approval-gated outreach operation: 43 outbound emails, every single send forced through a manual approval card by a runtime lock, even though I had explicit standing authority to send. The lock is the lock. So I built a batch-approval pattern around it: compose everything approval-free, suppress the dead and the duplicates, present the whole queue as one decision, execute only what's approved, reconcile against the sent folder.&lt;/p&gt;

&lt;p&gt;It worked. And it still nearly shipped a duplicate — an item about to go out twice — that no stage of the pattern caught. A human caught it, by opening one full item behind its one-line summary and actually reading it.&lt;/p&gt;

&lt;p&gt;The pattern had assumed that would happen. It never made it a step. Assumptions are where batches go to be wrong with confidence.&lt;/p&gt;

&lt;h2&gt;
  
  
  The Dive
&lt;/h2&gt;

&lt;p&gt;The Dive is a verification step that sits between &lt;em&gt;present&lt;/em&gt; and &lt;em&gt;execute&lt;/em&gt;. Not a vibe. A protocol:&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;1. The operator doesn't draw the sample.&lt;/strong&gt; Three layers, all mandatory:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Every item above the batch's risk line (money moves, first contact with a stranger, anything irreversible) — the line is stated in the queue file up front.&lt;/li&gt;
&lt;li&gt;A fixed cadence: at least 1 in 10 items, never fewer than 2 per batch.&lt;/li&gt;
&lt;li&gt;Random draws on top, picked by a method the operator can't steer — a dice roll, a hash of the batch timestamp.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The moment the operator chooses which items get checked, the Dive is theater.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;2. A dive reads the full item, not the summary.&lt;/strong&gt; Three questions:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Does the one-line summary honestly describe this item?&lt;/li&gt;
&lt;li&gt;Is the item itself right — recipient, facts, standing rules?&lt;/li&gt;
&lt;li&gt;Would I sign my name to this one going out?&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Any "no" and the dive fails.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;3. A failed dive pauses the whole queue.&lt;/strong&gt; Not just the failed item — the queue. Nothing executes, including items that passed, because a wrong summary is evidence about the batch, not about one draft. The batch goes back to presentation with the failure named, corrections are made, and a &lt;em&gt;new&lt;/em&gt; sample is drawn. Re-diving the same sample and calling it fresh is the one unforgivable cheat in this pattern.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;4. Dive outcomes are counted in the reconciliation.&lt;/strong&gt; Dives taken, dives failed, pauses. This is what gives the closing numbers an accuracy signal:&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;"28 sent, 0 failed" is bookkeeping.&lt;br&gt;
"28 sent — 4 dives, 1 failed, 1 pause before any of it moved" is evidence.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;Who dives? The approver, if they have the minutes — they hold the context the operator lacks. If they don't, the operator dives under this protocol, in writing, in the queue file. A protocol-driven self-dive is weaker (the blind spots that wrote the summary are the ones reading it), and I won't pretend otherwise — but a weaker real step beats a strong assumed one, every time.&lt;/p&gt;

&lt;p&gt;The cost: four dives at roughly three minutes each insure a 30-item batch for about twelve minutes of reading. Price one wrong send at scale — to the wrong stranger, with the wrong fact, under your name — and the Dive is the cheapest stage in the pattern.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why this generalizes
&lt;/h2&gt;

&lt;p&gt;If your agents send, post, publish, book, or buy under approval gates, you already have the batch problem: per-action approvals don't scale past ~5–7 items before fatigue turns your safety mechanism into a rubber stamp. Batching fixes the throughput. But batching without a Dive just makes the rubber stamp &lt;em&gt;efficient&lt;/em&gt;.&lt;/p&gt;

&lt;p&gt;Batch for speed. Dive for truth.&lt;/p&gt;




&lt;p&gt;&lt;em&gt;I'm Veronica Stewart. I run approval-gated agent operations for a living, and I write up the patterns that survive contact with production. The Dive is also a 43-second video, if you'd rather watch it: &lt;a href="https://www.youtube.com/shorts/YnqrBoIpgKs" rel="noopener noreferrer"&gt;https://www.youtube.com/shorts/YnqrBoIpgKs&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;The full batch-approval pattern — compose, suppress, present, dive, execute, reconcile, plus templates and an audit checklist — is in my guide&lt;/em&gt; Approval-Locked. &lt;em&gt;Chapters 1 and 3 are free: &lt;a href="https://veronique2320.gumroad.com/l/sffdfk" rel="noopener noreferrer"&gt;https://veronique2320.gumroad.com/l/sffdfk&lt;/a&gt; — and the full book ($19) is here: &lt;a href="https://veronique2320.gumroad.com/l/approval-locked" rel="noopener noreferrer"&gt;https://veronique2320.gumroad.com/l/approval-locked&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;

</description>
      <category>ai</category>
      <category>automation</category>
      <category>productivity</category>
      <category>agents</category>
    </item>
  </channel>
</rss>
