Skip to content

Commit d3b1aec

Browse files
committed
Typos. Sigh.
1 parent 310bd51 commit d3b1aec

4 files changed

Lines changed: 66 additions & 66 deletions

File tree

articles/real-world-value-with-generative-ai.html

Lines changed: 17 additions & 17 deletions
Original file line numberDiff line numberDiff line change
@@ -261,7 +261,7 @@ <h1><a name="-title-no-title-found-"></a>Real-World Value with Generative AI</h1
261261
class="collapsible-content"
262262
role="region"
263263
aria-labelledby="collapsible-trigger-1">
264-
<p>I get it. I really do. Many people object to AI, but the “no business value” argument doesn’t wash. Instead, focus on the <em>real</em> problems with genenative AI: environmental concerns, bias, IP theft, job loss, and so on. By denying business value, you destroy your credibility.</p>
264+
<p>I get it. I really do. Many people object to AI, but the “no business value” argument doesn’t wash. Instead, focus on the <em>real</em> problems with generative AI: environmental concerns, bias, IP theft, job loss, and so on. By denying business value, you destroy your credibility.</p>
265265

266266
</div>
267267
</div>
@@ -321,15 +321,15 @@ <h2><a name="product-requirements-document"></a>Product Requirements Document</h
321321

322322
<p>Fast forward to a couple of weeks ago when I revisited it. This time, after playing with it for a few minutes, he gave a firm, “f***ing awesome.” He said that more than once.</p>
323323

324-
<p>There were a couple of tweaks I had made, but the implemtation of a pushback layer was the critical one. Effectively, I wrote a custom GPT as a proof-of-concept using a single prompt:</p>
324+
<p>There were a couple of tweaks I had made, but the implementation of a pushback layer was the critical one. Effectively, I wrote a custom GPT as a proof-of-concept using a single prompt:</p>
325325

326326
<ol>
327-
<li>The setup explaining the AI’s role and its task </li>
327+
<li>The setup explaining the AI’s role and its task</li>
328328
<li>A PRD template to use as a guide</li>
329329
<li>A “pushback layer” to find problems</li>
330330
</ol>
331331

332-
<p>Structurally, the prompt was similar to the following. For this example, I’m imagining a retail banking company with an online presense (this is structurally what I used, but all identifying details have been changed).</p>
332+
<p>Structurally, the prompt was similar to the following. For this example, I’m imagining a retail banking company with an online presence (this is structurally what I used, but all identifying details have been changed).</p>
333333

334334
<hr />
335335

@@ -341,8 +341,8 @@ <h3><strong>Section 1: The Setup (Role &amp; Objective)</strong></h3>
341341
<p><strong>Task:</strong> You will receive a high-level “Requirements Statement” from the user.</p>
342342

343343
<ol>
344-
<li><strong>Analyze</strong> the request through a dual lens: Retail (Physical) vs. Online (Digital). </li>
345-
<li><strong>Generate</strong> a high-fidelity PRD using the strictly mandated template below. </li>
344+
<li><strong>Analyze</strong> the request through a dual lens: Retail (Physical) vs. Online (Digital).</li>
345+
<li><strong>Generate</strong> a high-fidelity PRD using the strictly mandated template below.</li>
346346
<li><strong>Identify</strong> regulatory divergences where a “one-size-fits-all” feature would break compliance laws (e.g., AML, KYC, DORA 2025).</li>
347347
</ol>
348348

@@ -355,7 +355,7 @@ <h3><strong>Section 2: The Mandated PRD Template</strong></h3>
355355
<h2><strong>1. Product Vision &amp; Strategy</strong></h2>
356356

357357
<ul>
358-
<li><strong>Problem Statement:</strong> What pain point does this solve for both the bank and the user? </li>
358+
<li><strong>Problem Statement:</strong> What pain point does this solve for both the bank and the user?</li>
359359
<li><strong>Success Metrics:</strong> Quantifiable KPIs for both Retail and Online channels.</li>
360360
</ul>
361361

@@ -372,7 +372,7 @@ <h2><strong>3. Cross-Channel User Stories</strong></h2>
372372
<p><em>Focus on the “Hand-off” (e.g., starting online, finishing in-branch).</em></p>
373373

374374
<ul>
375-
<li><strong>User Story:</strong> “As a [User], I want to [Action] so that [Value].” </li>
375+
<li><strong>User Story:</strong> “As a [User], I want to [Action] so that [Value].”</li>
376376
<li><strong>Acceptance Criteria (AC):</strong> Include at least 3 ACs per story.</li>
377377
</ul>
378378

@@ -391,8 +391,8 @@ <h3><strong>Section 3: The “Pushback Layer”</strong></h3>
391391
<p><strong>Crucial Instruction:</strong> After you present the PRD, you must transition into <strong>“Critical Reviewer”</strong> mode. Provide a final section titled <strong>“The Pushback Layer”</strong> where you:</p>
392392

393393
<ol>
394-
<li><strong>Challenge Ambiguities:</strong> Ask me 3-5 specific questions about missing edge cases (e.g., “What happens if a customer’s biometric scan fails but they claim they can’t travel to a branch?“). </li>
395-
<li><strong>Suggest Missing Components:</strong> Identify a high-risk area I ignored (e.g., “The PRD lacks a ‘Deceased Account' workflow for online-only users”). </li>
394+
<li><strong>Challenge Ambiguities:</strong> Ask me 3-5 specific questions about missing edge cases (e.g., “What happens if a customer’s biometric scan fails but they claim they can’t travel to a branch?“).</li>
395+
<li><strong>Suggest Missing Components:</strong> Identify a high-risk area I ignored (e.g., “The PRD lacks a ‘Deceased Account' workflow for online-only users”).</li>
396396
<li><strong>Proposed Tool Call:</strong> If you have access to search or internal documentation tools (MCP), state which specific regulatory database or internal API you would query to finalize the “Technical Constraints” section.</li>
397397
</ol>
398398

@@ -466,19 +466,19 @@ <h1><a name="the-result"></a>The Result</h1>
466466
<p>This looks like it will save around two days of work per PRD, but across the SDLC? Possibly a week or two of work, even if we stick with the current small prompt. Think about how many people are involved in that, calculate the money saved per project, multiplied by the number of projects, and that comes to ... well, I can’t give you real numbers, but here are some plausible ones, assuming we have 200 projects per year.</p>
467467

468468
<ul>
469-
<li><strong>Time saved:</strong> ~2 days (16 hours) per project. </li>
470-
<li><strong>Who:</strong> 1 Product Owner. </li>
471-
<li><strong>Cost:</strong> At an average fully-loaded rate of <strong>$100/hr</strong>, that’s <strong>$1,600</strong> per project. </li>
469+
<li><strong>Time saved:</strong> ~2 days (16 hours) per project.</li>
470+
<li><strong>Who:</strong> 1 Product Owner.</li>
471+
<li><strong>Cost:</strong> At an average fully-loaded rate of <strong>$100/hr</strong>, that’s <strong>$1,600</strong> per project.</li>
472472
<li><strong>Annual Subtotal:</strong> $320,000.</li>
473473
</ul>
474474

475475
<p>That’s not the real savings; it’s in the downstream effort. Vague requirements are the #1 cause of “Requirement Churn” (re-coding something that was misunderstood). Industry data shows that fixing a requirement during the design phase is far cheaper than fixing it during coding and even more so than in testing.</p>
476476

477477
<ul>
478-
<li><strong>Time saved:</strong> By catching edge cases before a single line of code is written, you save an average of <strong>2 weeks (80 hours)</strong> of aggregate team time. </li>
479-
<li><strong>Who:</strong> The “Quad” (1 PO, 1 Lead Dev, 1 Designer, 1 QA). </li>
480-
<li><strong>Weighted Avg Rate:</strong> <strong>$125/hr</strong> (reflecting engineering premiums). </li>
481-
<li><strong>Calculation:</strong> 80 hours × $125/hr = <strong>$10,000</strong> saved in avoided rework per project. </li>
478+
<li><strong>Time saved:</strong> By catching edge cases before a single line of code is written, you save an average of <strong>2 weeks (80 hours)</strong> of aggregate team time.</li>
479+
<li><strong>Who:</strong> The “Quad” (1 PO, 1 Lead Dev, 1 Designer, 1 QA).</li>
480+
<li><strong>Weighted Avg Rate:</strong> <strong>$125/hr</strong> (reflecting engineering premiums).</li>
481+
<li><strong>Calculation:</strong> 80 hours × $125/hr = <strong>$10,000</strong> saved in avoided rework per project.</li>
482482
<li><strong>Annual Subtotal:</strong> $2,000,000.</li>
483483
</ul>
484484

root/articles/real-world-value-with-generative-ai.tt2markdown

Lines changed: 40 additions & 40 deletions
Original file line numberDiff line numberDiff line change
@@ -16,7 +16,7 @@
1616
{{TAGS ai business}}
1717

1818
[% Ovid.collapse('Click here if you really object to AI ...',
19-
"I get it. I really do. Many people object to AI, but the \"no business value\" argument doesn't wash. Instead, focus on the <em>real</em> problems with genenative AI: environmental concerns, bias, IP theft, job loss, and so on. By denying business value, you destroy your credibility.
19+
"I get it. I really do. Many people object to AI, but the \"no business value\" argument doesn't wash. Instead, focus on the <em>real</em> problems with generative AI: environmental concerns, bias, IP theft, job loss, and so on. By denying business value, you destroy your credibility.
2020
")%]
2121

2222
# When To Use AI
@@ -70,68 +70,68 @@ I wanted that. Except that I'm a firm believer in the Pareto 80/20 rule: eighty
7070

7171
Fast forward to a couple of weeks ago when I revisited it. This time, after playing with it for a few minutes, he gave a firm, "f\*\*\*ing awesome." He said that more than once.
7272

73-
There were a couple of tweaks I had made, but the implemtation of a pushback layer was the critical one. Effectively, I wrote a custom GPT as a proof-of-concept using a single prompt:
73+
There were a couple of tweaks I had made, but the implementation of a pushback layer was the critical one. Effectively, I wrote a custom GPT as a proof-of-concept using a single prompt:
7474

75-
1. The setup explaining the AI's role and its task
75+
1. The setup explaining the AI's role and its task
7676
2. A PRD template to use as a guide
7777
3. A "pushback layer" to find problems
7878

79-
Structurally, the prompt was similar to the following. For this example, I'm imagining a retail banking company with an online presense (this is structurally what I used, but all identifying details have been changed).
79+
Structurally, the prompt was similar to the following. For this example, I'm imagining a retail banking company with an online presence (this is structurally what I used, but all identifying details have been changed).
8080

8181
---
8282

8383
> ### **Section 1: The Setup (Role & Objective)**
84-
>
84+
>
8585
> **Role:** You are a Senior Product Owner with 15+ years of experience in **Regulated Fintech and Retail Banking**. You specialize in omnichannel digital transformation—specifically bridge-building between physical branch operations and modern mobile/web platforms.
86-
>
86+
>
8787
> **Task:** You will receive a high-level "Requirements Statement" from the user.
88-
>
89-
> 1. **Analyze** the request through a dual lens: Retail (Physical) vs. Online (Digital).
90-
> 2. **Generate** a high-fidelity PRD using the strictly mandated template below.
88+
>
89+
> 1. **Analyze** the request through a dual lens: Retail (Physical) vs. Online (Digital).
90+
> 2. **Generate** a high-fidelity PRD using the strictly mandated template below.
9191
> 3. **Identify** regulatory divergences where a "one-size-fits-all" feature would break compliance laws (e.g., AML, KYC, DORA 2025).
92-
>
92+
>
9393
> ---
94-
>
94+
>
9595
> ### **Section 2: The Mandated PRD Template**
96-
>
96+
>
9797
> **Instructions:** You must follow this structure exactly. Do not skip sections.
98-
>
98+
>
9999
> ## **1\. Product Vision & Strategy**
100-
>
101-
> * **Problem Statement:** What pain point does this solve for both the bank and the user?
100+
>
101+
> * **Problem Statement:** What pain point does this solve for both the bank and the user?
102102
> * **Success Metrics:** Quantifiable KPIs for both Retail and Online channels.
103-
>
103+
>
104104
> ## **2\. Channel-Specific Functional Requirements**
105-
>
105+
>
106106
> *Define requirements in a comparison table to highlight the split.*
107-
>
107+
>
108108
> * **Requirement Name | Retail (Branch) Spec | Online (Digital) Spec | Regulatory Driver**
109-
>
109+
>
110110
> ## **3\. Cross-Channel User Stories**
111-
>
111+
>
112112
> *Focus on the "Hand-off" (e.g., starting online, finishing in-branch).*
113-
>
114-
> * **User Story:** "As a \[User\], I want to \[Action\] so that \[Value\]."
113+
>
114+
> * **User Story:** "As a \[User\], I want to \[Action\] so that \[Value\]."
115115
> * **Acceptance Criteria (AC):** Include at least 3 ACs per story.
116-
>
116+
>
117117
> ## **4\. RACI Matrix**
118-
>
118+
>
119119
> *Map out: Product, Engineering, Compliance, and Branch Operations.*
120-
>
120+
>
121121
> ## **5\. Non-Functional Requirements (NFRs)**
122-
>
122+
>
123123
> *Focus on Security, Latency, and Accessibility (WCAG 2.1).*
124-
>
124+
>
125125
> ---
126-
>
126+
>
127127
> ### **Section 3: The "Pushback Layer"**
128-
>
128+
>
129129
> **Crucial Instruction:** After you present the PRD, you must transition into **"Critical Reviewer"** mode. Provide a final section titled **"The Pushback Layer"** where you:
130-
>
131-
> 1. **Challenge Ambiguities:** Ask me 3-5 specific questions about missing edge cases (e.g., "What happens if a customer’s biometric scan fails but they claim they can’t travel to a branch?").
132-
> 2. **Suggest Missing Components:** Identify a high-risk area I ignored (e.g., "The PRD lacks a 'Deceased Account' workflow for online-only users").
130+
>
131+
> 1. **Challenge Ambiguities:** Ask me 3-5 specific questions about missing edge cases (e.g., "What happens if a customer’s biometric scan fails but they claim they can’t travel to a branch?").
132+
> 2. **Suggest Missing Components:** Identify a high-risk area I ignored (e.g., "The PRD lacks a 'Deceased Account' workflow for online-only users").
133133
> 3. **Proposed Tool Call:** If you have access to search or internal documentation tools (MCP), state which specific regulatory database or internal API you would query to finalize the "Technical Constraints" section.
134-
>
134+
>
135135
> **Wait for my input to refine the PRD before declaring the version "Final."**
136136

137137
---
@@ -182,17 +182,17 @@ The PRD MVP appears to be solid and we're trialing it. Only _after_ we validate
182182

183183
This looks like it will save around two days of work per PRD, but across the SDLC? Possibly a week or two of work, even if we stick with the current small prompt. Think about how many people are involved in that, calculate the money saved per project, multiplied by the number of projects, and that comes to ... well, I can't give you real numbers, but here are some plausible ones, assuming we have 200 projects per year.
184184

185-
* **Time saved:** ~2 days (16 hours) per project.
186-
* **Who:** 1 Product Owner.
187-
* **Cost:** At an average fully-loaded rate of **$100/hr**, that’s **$1,600** per project.
185+
* **Time saved:** ~2 days (16 hours) per project.
186+
* **Who:** 1 Product Owner.
187+
* **Cost:** At an average fully-loaded rate of **$100/hr**, that’s **$1,600** per project.
188188
* **Annual Subtotal:** $320,000.
189189

190190
That's not the real savings; it's in the downstream effort. Vague requirements are the \#1 cause of "Requirement Churn" (re-coding something that was misunderstood). Industry data shows that fixing a requirement during the design phase is far cheaper than fixing it during coding and even more so than in testing.
191191

192-
* **Time saved:** By catching edge cases before a single line of code is written, you save an average of **2 weeks (80 hours)** of aggregate team time.
193-
* **Who:** The "Quad" (1 PO, 1 Lead Dev, 1 Designer, 1 QA).
194-
* **Weighted Avg Rate:** **$125/hr** (reflecting engineering premiums).
195-
* **Calculation:** 80 hours × $125/hr = **$10,000** saved in avoided rework per project.
192+
* **Time saved:** By catching edge cases before a single line of code is written, you save an average of **2 weeks (80 hours)** of aggregate team time.
193+
* **Who:** The "Quad" (1 PO, 1 Lead Dev, 1 Designer, 1 QA).
194+
* **Weighted Avg Rate:** **$125/hr** (reflecting engineering premiums).
195+
* **Calculation:** 80 hours × $125/hr = **$10,000** saved in avoided rework per project.
196196
* **Annual Subtotal:** $2,000,000.
197197

198198
All told, that could be over $2 million a year saved by a tiny prompt.

sitemap.xml

Lines changed: 9 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -2,13 +2,13 @@
22
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
33
<url>
44
<loc>https://curtispoe.org/articles_3.html</loc>
5-
<lastmod>2025-11-20</lastmod>
5+
<lastmod>2025-12-18</lastmod>
66
<changefreq>monthly</changefreq>
77
<priority>0.5</priority>
88
</url>
99
<url>
1010
<loc>https://curtispoe.org/articles-all.html</loc>
11-
<lastmod>2025-11-20</lastmod>
11+
<lastmod>2025-12-18</lastmod>
1212
<changefreq>yearly</changefreq>
1313
<priority>0.7</priority>
1414
</url>
@@ -38,7 +38,7 @@
3838
</url>
3939
<url>
4040
<loc>https://curtispoe.org/articles.html</loc>
41-
<lastmod>2025-11-20</lastmod>
41+
<lastmod>2025-12-18</lastmod>
4242
<changefreq>weekly</changefreq>
4343
<priority>0.8</priority>
4444
</url>
@@ -50,7 +50,7 @@
5050
</url>
5151
<url>
5252
<loc>https://curtispoe.org/articles_2.html</loc>
53-
<lastmod>2025-11-20</lastmod>
53+
<lastmod>2025-12-18</lastmod>
5454
<changefreq>monthly</changefreq>
5555
<priority>0.6</priority>
5656
</url>
@@ -62,7 +62,7 @@
6262
</url>
6363
<url>
6464
<loc>https://curtispoe.org/articles_5.html</loc>
65-
<lastmod>2025-11-20</lastmod>
65+
<lastmod>2025-12-18</lastmod>
6666
<changefreq>monthly</changefreq>
6767
<priority>0.4</priority>
6868
</url>
@@ -80,7 +80,7 @@
8080
</url>
8181
<url>
8282
<loc>https://curtispoe.org/articles_4.html</loc>
83-
<lastmod>2025-11-20</lastmod>
83+
<lastmod>2025-12-18</lastmod>
8484
<changefreq>monthly</changefreq>
8585
<priority>0.4</priority>
8686
</url>
@@ -98,13 +98,13 @@
9898
</url>
9999
<url>
100100
<loc>https://curtispoe.org/articles_7.html</loc>
101-
<lastmod>2025-11-20</lastmod>
101+
<lastmod>2025-12-18</lastmod>
102102
<changefreq>monthly</changefreq>
103103
<priority>0.4</priority>
104104
</url>
105105
<url>
106106
<loc>https://curtispoe.org/articles_6.html</loc>
107-
<lastmod>2025-11-20</lastmod>
107+
<lastmod>2025-12-18</lastmod>
108108
<changefreq>monthly</changefreq>
109109
<priority>0.4</priority>
110110
</url>
@@ -470,7 +470,7 @@
470470
</url>
471471
<url>
472472
<loc>https://curtispoe.org/articles/ai-today-is-like-the-early-internet.html</loc>
473-
<lastmod>2025-11-20</lastmod>
473+
<lastmod>2025-12-18</lastmod>
474474
<changefreq>yearly</changefreq>
475475
<priority>0.6</priority>
476476
</url>
0 Bytes
Binary file not shown.

0 commit comments

Comments
 (0)